1. Research
  2. Products & Topics
  3. Periodicals
  4. Konzept
June 24, 2015
Few topics mirror the distortions in global economics and finance more than negative interest rates, share buybacks and changing corporate governance. We reflect on each in this latest issue. [more]
Konzept Issue 05 Reflections on unusual timesJune 2015 Few topics mirror the distortions in global economics and finance more than negative interest rates, share buybacks and changing corporate governance. We reflect on each in this latest issue. Cover story Reflections on unusual times— negative rates, buybacks and changing corporate governance happen when it ends. But it cannot be normal that some floating rate mortgages in Portugal turned negative, with the real prospect of borrowers receiving money each month from their bank. Being paid to own a house! Nor can economists agree on whether negative rates would have happened without quantitative easing anyway. What we do know for sure is that unprecedented distortions abound in part because an entire generation of investors has only known ever-lower borrowing costs. For instance, how else to explain record high western equity markets buoyed by share buybacks (that in theory add no value) and elevated profit margins (that in practice mean- revert)? Or, as hedge fund titan Jeff Gundlach recently noted, that the high-yield corporate bond market has never experienced a secular bear market? (Indeed the post-crisis years from 2010-2014 had the lowest period for defaults in modern history.) Try to justify the insane property prices in London, New York, Canada or Sydney? Or Chinese stock prices that have doubled within the last few months? You can’t. Or those technology companies, such as Instagram, valued in billions of dollars without as much as a cent of revenue? And how about Mexico’s 100- year government note with a 4.2 per cent yield to maturity? If any of the above was considered possible at a dinner party at the beginning of my career, guests would have fallen on the floor laughing. Yet now we talk about such things every day with barely a reference to how unusual they are. And if only unprecedented indicators stopped there. The roll-call also includes the developed world living under record-high levels of debt, with twenty advanced economies suffering private debt above 200 per cent of output; the VIX index of volatility within touching distance of historic lows; and slowing international trade as a proportion of global output. Nor has anyone ever before seen an economy as large and populous as China’s expand as quickly and for so long. Again the numbers are beyond comprehension. Humming along at double digit growth rates just before the financial crisis the country was using 23bn tonnes of natural resources a year, four times as much as the US, the second biggest consumer. Meanwhile, the global imbalances I have spent most of my academic and professional career – first at the universities of Harvard, Princeton and Chicago, followed by the International Monetary Fund and now chief economist of Deutsche Bank – thinking about the global economy and making predictions about it. Throughout these years, even as crises arose or when my forecasts were wide of the mark, the world never seemed too unfamiliar from what I had learned in text books or understood of history. Not so in these extraordinary times. As reflected in this latest issue of Konzept, it is no exaggeration to say we are living through one of the most unusual economic and financial periods ever. The scary thing to me is how rarely anyone seems to acknowledge this fact. From economics and markets to geopolitics and demography the world seems to have entered a distorted universe. And yet my beloved profession goes about its business making forecasts as if European sovereign bond yields, for example, were normal – not near their lowest levels in five hundred years. Just stop and think about that for a minute. Not their lowest ‘in decades’ or ‘since the second world war’ or ‘in a century’. Heavens, twenty generations have never seen interest rates at these levels! Even so economists discuss the current situation using the same old language and theories. Investors watch their screens as if nothing out of the ordinary is going on. The collective denial is unnerving. In America, Europe and Japan we pick apart quarterly data releases or the tiniest details of a central banker’s speech instead of stepping back to gasp in awe at arguably the biggest monetary experiments in history. Economists deal everyday in trifles such as: Where are Janet’s dots? Is a weak euro helping growth? Did core inflation rise a smidgeon in Japan? Yet the answers to such questions are immaterial next to $8tn-worth of quantitative easing (and rising) since the financial crisis. That is almost half the size of America’s economy in new assets on central bank balance sheets. And despite pretending otherwise no one really has a clue about the impact quantitative easing is having now, let alone what may Editorial on the UK’s membership of the European Union by 2017. None of these geopolitical uncertainties, still less the fact that a multitude of assets and indicators are trading at either historic highs or lows, will stop economists from trying to predict the future. You might think they would know better only eight years on from failing to foresee the biggest crisis since the great depression and barely three years since many predicted Greece’s imminent exit from the eurozone. Whichever way this unusual state of the world eventually pans out, I can guarantee that no one will have read an accurate description of it anywhere beforehand. When thinking about such things, I am reminded it was only 30-odd years ago that the French godfather of mathematical economics, Gerard Debreu, won the Nobel Prize for his work modelling abstract economies in order to prove the existence of a general equilibrium. What surprises me today is not how much the current state of the world differs from these abstract, unreal economies – that is obvious enough to everyone. Rather I’m amazed at how far removed we are from what the subsequent critics of Debreu’s work consider to be a more realistic view of the world. David Folkerts-Landau Group Chief Economist Member of the Group Executive Committee caused by China’s unprecedented accumulation of savings have affected everything from inflation in Portugal to the share price of Louis Vuitton. These imbalances remain with us today and also explain why the global financial architecture is changing before our eyes. New Chinese-led multilateral institutions such as the Asian Infrastructure Investment Bank were inconceivable not so long ago, as was full renminbi convertibility or its inclusion in the basket of currencies that make up the IMF Special Drawing Rights. What is more, if a relentlessly rising China made the world an unrecognisable place, do not expect things to become easier now that the country is cooling. A slowdown on this scale also is unprecedented. The reality is China’s working age population is shrinking as a proportion of the total and a historic mass migration from rural to urban areas seems to have run its course. Thanks to the lure of the city, China has almost the same number of migrant workers as America has people. Now only about a fifth of the country’s labour force is left working in agriculture. This ever smaller rural supply of labour, together with China’s rising dependency ratio since 2013 (births peaked in 1987 due to the one child policy), will accelerate wage growth and in the process hurt profits and investment. Of course such a shift is the rebalancing economists have expected and long wished for. But that does not make the global repercussions of such a monumental shift in China’s economy any easier to predict. Throw in India, another country with a billion-plus people, with its desire to emulate China’s progress of the last twenty years and in my opinion anyone making long-run global economic forecasts these days is deluded. So much uncertainty and I have not even begun to mention geopolitics. How do economists input into their spreadsheets the tensions over control of the South China Sea, Russia’s revanchism and cold war attitude towards the west, or the fractures within the eurozone? What is the right energy price to use when two great Middle Eastern powers are becoming increasingly belligerent across the region, albeit mostly from behind the scenes? Or try modelling the UK economy knowing that there is now definitely going to be a referendum To send feedback, or to contact any of the authors, please get in touch via your usual Deutsche Bank representative, or write to the team at research.haus@db.com. Articles 06 Cloud computing—a grounded response 08 Leases accounted—behold more hidden debts 10 From concentrate—America’s diluted competition 12 German retail banks—at home with overseas customers 14 Currency disintegrations—picture an exploding bomb 16 Inflation targeting—the downside of aiming higher Columns 68 Book review—the chimp paradox 69 Ideas lab—the threat of artificial intelligence 70 Conference spy—dbAccess Asia 71 Infographic—corporate America: 1995-2015 German model—has a consensus economy reached its limit? 54 European capital markets— living up to the weight of history 60 Features One share, one vote— one mess 36 Share buybacks—truths, myths and technical bits 19 Work crisis—a divided tale of labour markets 46 Credit freak show— money for nothing 30 Cloud computing— a grounded response Bryan Keane, Karl Keirstead, Nandan Amladi at 30 per cent compound each year, a rate that is expected to continue for at least the next three years. Some of the world’s biggest technology companies including Google, Microsoft, and IBM have scrambled to take advantage. These impressive figures, though, come from a very small base. Indeed, the $150bn or so that businesses are currently spending on cloud computing represents just five per cent of all technology spending. It seems that business managers have been reluctant to try and emulate Nasa’s success. Several factors appear to be holding back the decision to migrate to the cloud. First, technology outsourcing proposals frequently fail the so-called wince test. Unlike labour arbitrage that can be achieved by moving, say, factories offshore, and the “lift and shift” of entire non-core business functions such as human resources or logistics to external service providers, outsourcing technology involves a far greater transfer of trust. That is, control of intellectual property and sensitive data as well as responsibility for ensuring minimal down-time shifts outside the organisation’s home base. This can be particularly unnerving for mission-critical systems and those where latency is an issue. It is perhaps no surprise, then, that although about one-fifth of companies use cloud-based services for some human resources work, just one-tenth use them for their sales and procurement process, half that figure for legal, supply chain, and logistics work, while just two When the Curiosity Rover descended onto the Martian surface in 2012, the 80,000 requests per second for live video streaming that hit Nasa’s website consumed a colossal amount of computing power. That muscle was provided by Amazon, the world’s biggest cloud computing provider. Outsourcing its processing requirements to the cloud saved Nasa the expense and hassle of buying and maintaining a huge bank of servers that would only be needed to deal with infrequent demand spikes. At the time, Nasa’s successful use of the cloud seemed to be the best possible advertisement for both the technology and the burgeoning industry growing up around it. The idea was, and still is, a simple one: an organisation can have someone else process or store data rather than doing it in-house – something that can be particularly useful for organisations that experience the occasional but significant surge of customer demand. From one perspective, in the three years since Curiosity began rolling around the red planet, cloud computing has taken off, growing Konzept 6Konzept 6 per cent of firms trust cloud providers with their marketing function. It is true that many cloud storage and processing companies have greater levels of security than the existing data centres of their clients but the digital world is littered with security systems once thought infallible and there is a psychological impact to the chief technology officer of sensitive data being hacked through someone else’s system rather than company’s own. The second issue is career risk for the chief technology officer. Moving an organisation’s systems to the cloud requires a significant change program and lead times. It may also involve writing off many of the organisation’s existing assets, costs that quarterly-result conscious executives may wish to avoid. Third, once outsourced, an organisation is somewhat locked-in to the cloud provider. This does not necessarily mean the vendor will hold its clients’ data hostage – legal contracts can prevent that – but rather it must be assumed that there will be unknown future infrastructure transfer costs. The fourth issue concerns the way technology expenses appear in financial statements. Traditionally, when a company buys a bunch of servers they appear as assets on the balance sheet and a depreciation charge is put through the income statement each year. Crucially, this does not affect ebitda – a figure that features prominently in senior management’s performance assessment, not to mention a company’s debt covenants. Under an outsourcing arrangement, the amounts paid to an external provider would appear as an operating expense and thus decrease ebitda. Many of the industry’s pioneers thought the cost savings promised by the cloud would make migration irresistible for clients. Surveys show, however, that even companies that use a “private cloud”, which restricts access to certain users within an enterprise, will only do so if they feel it will increase productivity. Indeed, just three per cent of companies set up a private cloud in order to reduce costs. So if cost savings are not the driving force for the adoption of cloud computing, should technology managers only bother about it if their business is susceptible to spikes in website or systems demand? One thing that may change that perception is big data. Despite being a term without a concrete definition, it is quickly becoming a focus point for companies. Consumer-driven firms such as Amazon and Walmart are just two that are discovering how to use their vast reams of information; more will follow. Most companies, though, simply cannot store the quantity of data they wish to collect nor justify the cost of purchasing the hardware to process it all. This is why cloud computing may not only come in handy in the near future but may even become mandatory. Currently, about five per cent of processing is carried out in the cloud but over the next ten years this could increase to between one-quarter and one-half. Much of this increase may be driven by a relatively low number of high-effort big data requests rather than a very high number of low-effort requests. Big data is still in its infancy so it is difficult to predict how it will mature. And of course, what qualifies as big data changes over time. In 1969, the Apollo 11 guidance computer overloaded during the descent to the moon. Today, the average smartphone could handle the job as the ubiquity of the internet has fuelled demand for storage space and processing power. So while it is tempting to think that the relatively sudden growth of cloud computing may peter out as companies continue to purchase storage and processing power ever more cheaply (think Moore’s law) it is far more likely that organisations will continue to do what they have always done – consume technology resources as fast as they come online. The pace at which organisations migrate to the cloud will vary, by firm and country. Aside from cost and demand, regulatory considerations that include data sovereignty, security, and auditability will also determine the pace of technology adoption. Individual companies will also weigh up their size and potential scale benefits. Layers of their technology “stack” that can be most usefully outsourced depend on the extent of customisation needed. In other words, the extent to which an organisation’s technology will work on someone else’s machine. So far, the vast majority of cloud-based infrastructure uses open-source technology that makes it easier for different systems to talk to each other. That bodes well for large providers like Amazon, Google et al, which are plunging seemingly endless resources into becoming cloud providers, not to mention the plethora of niche competitors. Many of these cloud vendors will not survive, especially if the industry becomes commoditised and prone to low margins. But that will only happen if the entire market for cloud computing expands dramatically. If that is the case, then the cloud is set to increase its influence over every organisation. Please go to gmr.db.com or contact us for our in depth reports: “The Data Chasm” and “Digital Changing IT Services Model” Konzept 7 Here is a terrifying game of ‘What if?’. What if you were chief executive of a large company and one morning your accountant says your debt is one-quarter higher than it was yesterday, and debt to assets and debt to equity ratios have ballooned one-fifth and one-quarter respectively? Actually this is no game. These are real estimates by American and international accounting bodies for when, in just a few months’ time, they force companies to bring leases on-balance sheet. Moving a lease on-balance sheet means adding up all future payments, whether they are for a building, a car, or even a photocopier, and adding the result (discounted to present value) to the liabilities on the company’s balance sheet. Hey presto! Extra debt. This accounting change will affect almost every company almost everywhere in the world and the numbers are eye-popping. A survey of 1,000 large global firms showed that the 25 per cent extra liabilities equates to $1.5tn – about four times the public debt of Greece, all currently hiding off-balance sheet. And that is just the companies in the survey. You may dismiss this as purely cosmetic – all numbers on a page with no actual business impact. That would be a mistake. Here are five very real implications. First, interest expense will rise for non-US firms. This occurs because the interest component of each lease payment will now be split out and shown for what it is, a financing cost, rather than being buried inside the ‘lease expense’ line as it is presently (different rules will apply in the US). Second, because interest payments tend to be front loaded in a series of loan payments, the additional interest charge on a company’s profit and loss statement will be higher at the beginning of the lease and lower at the end. Third are the flow-on effects, particularly on many ratios that form part of loan covenants. Take, for example, a company with a 20-year lease on a building. Assume an annual lease payment to its landlord equivalent to about one- tenth of revenues, a return on assets of 10 per cent and a net debt to ebitda ratio of 3.5. Under Leases accounted —behold more hidden debts Luke Templeman Konzept 8Konzept 8 the new rules, when it brings the lease onto the balance sheet, its return on assets will fall two- thirds, interest cover will halve and net debt to ebitda will double. Fourth, the cost of borrowing will almost certainly rise for some companies. Even the rule-makers admit this. This is because creditors will now have clarity on the extent of a company’s future liabilities – something they have not had before. No doubt some companies will experience an ‘Enron moment’ as a flood of liabilities that had previously been lurking off-balance sheet are brought into the spotlight. Brace for a knee-jerk response from lenders. Fifth, and finally some good news for companies: negotiating leases may just become a little easier. In the past, some treasurers were willing to accept more onerous terms from lessors to ensure that a lease contract met the definition of an ‘operating lease’ and thus kept off-balance sheet (as opposed to a ‘finance lease’ that is kept on). Now that all leases will be treated equally, lessees may be able to drive a harder bargain. Why go through all this upheaval at all? Most obviously, international accounting bodies want to force companies to disclose exactly what their liabilities are. It is true that some analysts already use a rule of thumb and multiply a company’s lease expense by eight to approximate its total lease liabilities. But this approach is a finger-in-the-wind estimate. The IASB thinks this old trick actually overstates the real liability by almost one-tenth. The other main aim is to make it easier for laymen to professionals to compare different companies. Under the current rules, two identical firms with identical leases can have very different balance sheets if one puts leases on balance sheet while the other keeps them off. Geographical anomalies will also be normalised. For example, the IASB reckons liabilities of large companies are understated by one-fifth in Europe, one-quarter in the US and almost one- half in Asia. Soon this should not be an issue. This all sounds great but there is an easily- missed quirk that could make life difficult for both investors and treasurers. It concerns the discount rate that is applied to the series of lease payments to determine their present value. In most cases, the rate used will be fixed when the lease is signed and will usually be the company’s marginal borrowing rate. The problem arises as interest rates change. From an investor’s point of view, two identical companies with identical lease obligations can still have a different balance sheet if they signed their lease contracts at different times or in different countries with different interest rates. From a treasurer’s viewpoint, the prevailing interest rates in the market, or an expected upgrade or downgrade of the company’s credit rating may affect when, or in what jurisdiction, a lease can be signed and on what terms. To illustrate, once again consider the hypothetical company mentioned above. Say it signs its 20-year lease today. The super low interest rates available may translate into a cost of borrowing of, say, four per cent and the company would discount its expected lease payments accordingly. If in the future, however, another identical company signs an identical lease, but at a time when base rates have risen by four percentage points, it would have a cost of borrowing of eight per cent. The mathematics of the discount process mean the second company’s liability would be about one-quarter lower than that of the first company even though the cash leaving the door is identical. Not to mention that the second company’s net debt to ebitda and interest cover ratios would both be one-fifth lower than those of the first, while return on assets would be about one-fifth higher. Weird side effects then become apparent. Just one is that the worse a company’s credit rating, the lower its apparent liability. The effects of the new rules may take many companies by surprise but the accounting bodies cannot be accused of rushing the rules out – the initial report for discussion was published in 2000. Fifteen years have since passed and the final announcement comes not a moment too soon. Given the sheer size of the liabilities involved, it is easy to imagine a parallel universe in which a company with heavy off-balance sheet lease commitments goes bust and saddles investors and creditors with more liabilities than they knew existed. In that situation, new rules would have been rushed into place without the extensive consultation process that standard setters have undertaken. This happened after Enron’s demise when, arguably, some of the subsequent rules were too hastily implemented and did not achieve their aims. So hats off to the US and international accounting bodies in this case for taking the time to construct carefully considered rules in close collaboration with companies, advisors, and end users. Other regulators should take note. Konzept 9 First, the lure of record-high profits should be spurring on new business start-ups. Instead, for all the dynamism of Silicon Valley, over the last generation, America has witnessed a trend of declining entrepreneurship. Firm entry rate (new businesses formed each year as a proportion of the existing number of total businesses) has been declining for over 30 years and has halved from nearly 15 per cent in the late 1970s to eight per cent more recently. 1 Along with declining start-up rates, consolidation among existing players is the other big force working to reduce competition. In the mid-1980s, the US issued a revised set of merger guidelines that loosened the criteria used to evaluate consolidation deals from an anti-trust perspective. What has resulted is an increase in industry concentration across a whole host of sectors. Census data show the proportion of manufacturing sectors where the top-four firms control over half the market rose from 30 per cent in 1992 to 40 per cent in the latest available dataset for 2007. Take US airlines, for example, with its chequered history of delivering profits and many high-profile bankruptcies. The industry is currently going through something of a purple patch with $12bn in net profits last year, compared with $3.5bn in 1999. This three-fold increase in profits was accompanied by the top-four US carriers increasing their market share of domestic air travel from 60 per cent to 85 per cent. On individual routes, the consumer choice is even more restricted. Whereas 25 years ago, passengers flying from New York to Cleveland could choose from nine or ten different airlines, today they are restricted to three big carriers – and only one if they prefer flying out of Newark airport. The fortunes of their European counterparts are in stark contrast. Despite sporadic merger deals, Europe’s skies remain full of economically unviable flag carriers and competition is fierce. The four biggest airlines have, between them, less than 40 per cent share of intra-European travel. Indeed the 25 largest airlines in Europe have a combined 85 per cent market share, the same as controlled by just the top-four in the US. Why does this matter? A more consolidated US industry has been more restrained in capacity expansion with total available seat miles falling over the last decade. In contrast, European capacity is 75 per cent higher over the same time period. The impact is visible in pricing as well. Average US domestic air fares are 20 per cent higher today than in 1999. While equivalent aggregate European data is not available, the average annual air fare for the low-cost carrier Ryanair has actually declined over this period. While many characterise the current era as the “consumer age”, in reality it is the golden age of the American corporation. The combined earnings for S&P 500 companies topped $1tn last year and their cash vaults strain with $2tn of gross liquid assets. Profit margins have soared to multi-decade highs and stock market valuations once again approach previous cyclical peaks. In the quarter of a century since 1990, corporate earnings have increased at a seven per cent annualised rate, comfortably outpacing the four per cent nominal output growth. Consequently, the share of US corporate profits in the country’s output has doubled over the last 25 years to the highest in post-war history. Corporate America has multiple sources to thank for this bonanza. The seemingly obvious ones are the falling costs of both labour and capital. The availability of cheap and plentiful workers in a globalised world as well as record-low interest rates has helped lower overall costs for businesses. However, the idea that falling costs alone can lead to higher profit margins, borders on economic illiteracy. In a well-functioning competitive market these cost-advantages should be quickly passed on to end consumers. Of course the words “competitive market” are key in the assertion above. The story of American business over the last generation, however, is one of declining competition. From concentrate— America’s diluted competition Rineesh Bansal Konzept 10 Konzept 10 Is it any wonder that the eight per cent ebit margins for US commercial airlines are higher than any other region and three times more than in Europe? Telecommunication service providers tell a similar story. After three decades of industry consolidation, the 300m strong US mobile phone subscriber base has come to be shared among only four major players. Since then, however, American regulators have thwarted attempts to shrink the sector down further to three providers by objecting to the acquisition of T-Mobile USA by both AT&T and Sprint. Contrast this with telecoms in Europe, where three or four competitors exist in each of the relatively small 28 national markets. Stéphane Richards, the chief executive of Orange, is on record saying “It is crazy that there are over 100 fixed and mobile operators in Europe”. More recently, European regulators have viewed consolidation deals in the telecom sector more favourably. However, the European Commission’s website sums the situation quite well stating: “Despite rather competitive national mobile markets, a full functioning single market for mobile communications does not yet exist.” Does this preponderance of service providers help or hurt European consumers? On average, European consumers pay about half as much as Americans for their monthly mobile bills, although this also reflects the significantly higher voice and data usage in the US 2 . However, the result of fragmented European markets is that even the smallest of the four big American mobile carriers has more subscribers than the largest European service provider. If the lack of scale among European players creates a hurdle for required investment, consumers will suffer the consequences in lower quality services. A similar scenario unfolds in banking. Even though both the US and euro area have about 6,000 banks each, America is a large single market dominated by a few large institutions while European banks remain constrained by national idiosyncrasies. The proportion of total US bank assets held by the top-five banks increased from one-quarter in the late 1990s to nearly one-half currently. For Europe the ratio still stands at just above one-quarter. 3 Hence, if large bank concentration in the US was restricted to European levels, America’s biggest lenders would have to halve in size. How much advantage does this bestow upon US banks? The Federal Reserve estimates this would raise the ratio of non-interest costs to operating income of US banks by three to six percentage points. 4 The American consumer is an economic force unrivalled in size and strength. For US companies, ready access to this huge pool of reliable and homogenous demand is the source of an enormous structural advantage that allows them to prosper at home and dominate abroad. Alongside the thriving financial performance of businesses, the last quarter of a century has also witnessed an increasing trend of a few players commandeering ever-greater control of their respective industries. Peter Thiel, the co- founder of Paypal, perhaps summed up corporate America’s sentiment in the title of his Wall Street Journal article ‘Competition is for losers’. Left unchecked, this tendency towards decreasing competition threatens eventually to hurt the very consumers that are the underlying strength of American business. European companies, on the other hand, face-off against their American peers on a lopsided playing field. Rather than enjoy the benefits of operating in a single market with 600m consumers, far too often they are hampered by having to deal with 28 small markets segregated by national boundaries. Some nascent moves like the single European banking regulator reflect progress towards removing existing structural impediments. Similarly, recent deal activity – both cross-border and intra-country consolidation in banking and telecom – holds promise. Were these trends to continue and accelerate, we could even talk of a golden age of European companies. 1 Brookings Institute, “Declining Business Dynamism in the United States”, May 2014, Ian Hathaway and Robert Litan 2 “Mobile Wireless Performance in the EU & the US”, GSMA Intelligence 3 “International Comparison of Banking Sectors”, European Banking Federation, data as of end-2011 4 “Do big banks have lower operating costs”, New York Federal Reserve Economic Policy Review, December 2014 Konzept 11 cent. Furthermore, it is estimated that migration will raise the potential economic growth for the coming years by 0.4 percentage points. The increased labour supply, nonetheless, slightly dampens the rise in wage levels. 2 While Germany’s migrant community is not a homogenous group, some common characteristics are worth highlighting. For example, with an average age of 35 years, the migrant group is 12 years younger than the native German population. 3 Studies have also shown that migrants are more likely to become entrepreneurs. Hence, immigration is helping alleviate Germany’s demographic challenge and securing the country’s future economic strength. But what about the investment preferences of migrants? In cooperation with the University of Bayreuth, we conducted a study analysing the differences in risk attitudes of German natives and migrants. Using German socio-economic panel data collected by the DIW research institute in Berlin which covered 30,000 citizens, we scrutinised the general attitude towards risk as well as the willingness to take risk when it comes to financial investments. 4 For this purpose, migrants were defined as inhabitants not in possession of German citizenship. While sweeping generalisations about the migrant group should be avoided, the detailed findings in our study still offer some noteworthy results. An individual’s place of birth, for instance, has an important impact on their general risk attitude. Not only do migrants born abroad show a greater risk aversion than German nationals, but they are also significantly more risk averse than their fellow migrants who were born in Germany. German retail banks— at home with overseas customers Germany is in the midst of an immigration boom. As the immigrant population swells to become a sizable customer group, gauging their preferences will be increasingly important for German businesses in all industries. Banks and financial service providers, for instance, should understand if the investment style of immigrants varies significantly from the native population. This article aims to shed some light on this hitherto less researched topic. Germany welcomes more immigrants than any other country in the European Union. And even among the OECD countries only the US accommodates more. In 2013, more than 1.2m people moved to Germany, a two-decade high. Net migration that year numbered over 425,000. Among those moving to Germany, three-quarters originate from Europe and two-thirds are European Union nationals. Migration statistics for 2014 are expected to outpace the figures for 2013. In Germany today, almost 17m inhabitants are considered to have a migrant background. 1 With such manpower, migrants have become an important driver of the German economy. Between 2010 and 2013, one-fifth of the economic performance can be attributed to the migrant labour force. In fact, without immigrants, the German economy would have shrunk in 2014 instead of growing by 0.4 per Nicolaus Heinen, Lea Bitter, Timo Alberts (University of Bayreuth) Konzept 12 Konzept 12 1 BAMF (2015). Migrationsbericht des Bundesamtes für Migration und Flüchtlinge im Auftrag der Bundesregierung (Migrationsbericht 2013), Bundesministerium für Migration und Flüchtlinge. 2 Folkerts-Landau, David (2014). Temporary immigration boom: A wake-up call for politicians? Deutsche Bank Research Standpunkt Deutschland, Frankfurt. July 2014. 3 Statistisches Bundesamt (2014). Bevölkerung und Erwerbstätigkeit, Bevölkerung mit Migrationshintergrund – Ergebnisse des Mikrozensus 2013. Fachserie 1, Reihe 2.2, Wiesbaden 2014. 4 DIW (2015). SOEP Overview. Berlin: Deutsches Institut für Wirtschaftsforschung. Website: http://www.diw.de/en/ diw_02.c.222508.en/soep_overview.html 5 Deutsche Bank (2015) Bankamız – das Angebot für türkischstämmige Kunden in Deutschland. Frankfurt am Main, p. 18 However the picture changes when looking specifically at the risk attitude towards financial investments. In this case, all migrants are slightly more risk tolerant than German natives. Analysing the data along gender lines indicates that whereas female citizens tend to be significantly more risk averse than men, female migrants have a lower risk aversion than the average native German woman. Finally, we analysed the financial behaviour of individuals who assessed themselves as particularly risk loving. Consistent with our previous findings, risk-loving migrants possess a significantly higher willingness to take risky financial decisions than their risk-loving German counterparts. The salient question for banks is how can such knowledge be leveraged in daily business? There are three practical implications for banks’ daily customer service. 1. Being aware of cultural differences is key. The knowledge that clients with a migrant background possess a different risk attitude is essential for custom-tailored advice and might prevent misunderstandings. 2. Enquiring about the general risk attitude can be crucial. Migrant customers with a high general willingness to take risk are also overly willing to take risk when it comes to financial investments. Considering a customer’s general risk attitude can therefore help the bank to recognise and offer them the most appropriate products. 3. Assisting the transformation of risk profiles guarantees a successful and sustainable customer relationship. For instance, since 2006, Deutsche Bank has offered special services under the name of Bankamız for migrants with a Turkish background. Experience shows that, with an increasing length of stay, first-generation migrants adapt their risk attitude towards that of mainstream society. In this context, considering the transformation of risk preferences is a crucial factor for offering advisory service and will improve the long-term customer relationship. The analysis is not just for daily retail banking. The findings also suggest three strategic opportunities. 1. Addressing target groups appropriately pays off. Retail banks that satisfactorily cater for the preferences of migrant customers can benefit from their broad network. When migrant customers are happy with their bank, they tend to recommend it more often than native customers do. Statistics reveal that the likelihood of Bankamız customers to recommend their financial service provider to a friend or acquaintance is almost two- thirds higher compared with natives. 5 These “recommendation returns” stem purely from responsible and diligent customer relations. 2. Awareness of untapped potential can unlock new customer groups. So far, major banks in Germany have focused mainly on Turkish migrants, the biggest group of German immigrants. Turkish migrants have a broad media network and thus are easy to address. But immigrants from Russia, Poland and former Yugoslavia are likewise promising target groups. 3. The market for financial services for migrants is growing. As mentioned above, migrants are a growing part of German society and their qualifications and incomes tend to increase the longer they live their host country. In addition, their demographic characteristics are more advantageous than those of native Germans. It seems that customers with a foreign background will gain increasing importance in the future. The only question is, which banks will be the most ready for them? If financial service providers incorporate these findings into their business model, tailor-made services for migrant customers will become a daily and rewarding business. The huge potential of citizens with a migrant background is easy to spot and – with a suitable approach – easy to leverage. For more details, please see our report “Clients with a migrant background: The role of risk preferences in retail banking”, Current Issues, on dbresearch.db.com Konzept 13 day comes. To this end, while history never repeats itself, much can be learned from the disintegration of Austria-Hungary in the autumn of 1918. The best metaphor when thinking about the inflation effect of a disintegrating currency area is an exploding bomb. The centre of gravity of the explosion continues along its original trajectory, even as pieces of shell fly off in multiple directions. The acceleration of some fragments will be matched by the opposite acceleration of others. Likewise, while the overall inflation in a disintegrating monetary union follows its original trajectory, inflation in individual countries can diverge rapidly. Hence in the case of Austria-Hungary, there was rapid and steady inflation prior to the disintegration of the unified crown zone. The Austro-Hungarian Bank was controlled by the interests of the dual monarchy, which was desperate to finance an existential war through an inflation tax. This policy was evident in the steady growth of notes in circulation at the time, as well as the currency depreciation versus the dollar prior to 1919. 1 Within months of the collapse of Austria- Hungary, however, the three successor country fragments flew off in radically different directions. For instance, once deflationary Czechoslovakia had printed sufficient new currency to replace the old crowns and satisfy internal, non-inflationary demand, it stopped increasing the currency supply. Significant Currency disintegrations —picture an exploding bomb However you perceive the prospects for Greece staying in the eurozone, a cursory view of history reveals that when it comes to currency area disintegrations, Europe has plenty of past form. Except for the post-world war two reorganisation of former European colonies into separate sovereigns, almost all the world’s currency area disintegrations in the past century have occurred in Europe, with the last one only 22 years ago. Of course this is not surprising and hence an unfair charge. Unlike in other continents, the frontiers and sovereignties of Europe have shifted frequently. The current eurozone is an impressive achievement and investors should wish it success and stability for many decades to come. But at the same time there is a disingenuous ring to it when policy makers insist that monetary union in Europe is “irreversible” or “irrevocable”. Hence it pays to remind ourselves of how currency areas disintegrations play out. Of most value to investors is understanding the potential for inflation and exchange rate movements in newly-separated states, if and when the Peter Garber Konzept 14 Konzept 14 1 See charts on page 26-27 in “The Dissolution of the Austro-Hungarian Empire: Lessons for Currency Reform”, 1994, Peter Garber and Michael Spencer, available on http://www.princeton.edu/~ies/IES_Essays/E191.pdf appreciation versus the dollar followed. In contrast, Austria and Hungary’s legacy central banks pushed currency printing towards hyperinflation, collapsing their exchange rates in the process. Why these divergent paths? Czechoslovakia emerged from the disintegration with a serious industrial economy and tax base and no reparations bill. Therefore it had no need to generate tax revenues via inflation. Austria, on the other hand, was a rump state left with an oversized capital city that had supported its grandeur by draining revenues from outside – particularly Czechoslovakia. It now had an inadequate tax base to maintain its accustomed expenditures. Indeed after the first world war both Austria and Hungary shared the tax revenues made available because of inflation, so it was natural for them to want rising prices upon separation. Czechoslovakia stopped this feast on its tax base by controlling the amount of currency already in internal circulation, closing its borders to prevent a further inward rush of Austrian currency, and starting its own conservative bank of issue. With Czechoslovakian revenues no longer available to them, Austria and then Hungary had to print ever faster to cover their still uncontrolled expenditure. Thus were born the famous post-world war one periods of hyperinflation in Austria and Hungary that economics students read about in textbooks. Eventually the League of Nations took control in order to put the two countries’ finances in order. In fact, these rescue programs became early templates for the rebuilding of collapsed sovereign finances that were institutionalised by the IMF a generation later. There is one other noteworthy lesson to draw from this historical example of currency disintegration: the country in a currency area that most dislikes having real resources drawn away from it by central bank operations is the one that will trigger the disintegration. If that country does not control the central bank, it will secede in order to start up its own, throwing up transitional capital controls and border closures to shut out the influx of further currency issuance from the old central bank. In relation to the current euro crisis, the country that seems to dislike having resources drawn away from it the most does appear to be in control of the central bank. Hence, were a disintegration to occur in order to stop such a transfer of resources, that country would likely inherit the central bank. Instead of the border closures in the Austria-Hungary example, ruptures are now done via the central bank’s payment system and by squeezing the scope of acceptable collateral in the targeted country, limiting new resources that can be extracted from the controlling group. So far, however, the likes of Germany have acquiesced in the outflow of resources via the European Central Bank. So what does our bombshell metaphor imply for eurozone countries in the event of a Greek ejection from the euro? In the Austria- Hungary example the successor countries were of similar weight, so the shifts in trajectory were large for each of them. However in the eurozone’s case, Greece is small so although the remaining members may have a bit more deflation and less currency depreciation after the initial shock dissipates, the shifts from their present paths should be minor. The story is different for Greece. As happened to Austria and Hungary in their unfriendly separation, Greece may be thrown onto its own resources while still trying to preserve some semblance of its pre-crisis income distribution. One big difference, however, and a reason for optimism, is Greece may have real outside support from the IMF. In the IMF’s absence there are also other geopolitical lenders of sufficient weight to catch Greece on the rebound, and these may assume an IMF-like role in a test of potential substitutes for the Bretton Woods institutions. That should help prevent unnecessary harm to Greece’s internal economy and international economic relationships until the dust settles after the explosion. Konzept 15 Inflation targeting— the downside of aiming higher Desperate times call for desperate measures. Imagine the developed market central banks and governments delivering a coordinated monetary and fiscal policy stimulus in order to push inflation to a new target of four per cent – two whole percentage points above current targets. Should they succeed in credibly establishing a new inflation environment, the price of 10- year government bonds would fall by about 15 per cent and 30-year bonds would plummet by more than a third simply on account of the corresponding increase in nominal yields. There would be a large redistribution of wealth from creditors to debtors. Calling this a shock to the system would be an understatement. Yet this is no complete fiction. In the aftermath of the financial crisis, there have been proposals to set inflation targets meaningfully higher in order to tackle the problem of the effective lower bound on nominal interest rates, which prevents central bankers from setting policy rates much below zero. 1 Higher inflation would allow real interest rates to fall lower in deep recessions, speeding up recovery. In fact, as early as 2010 the IMF suggested 2 that policymakers consider raising the inflation target from two to, say, four per cent, with the qualification that a careful cost-benefit analysis would be required. Such an inflation target increase could be done gradually to soften the blow. Frankly, it would have to be, given present policies are struggling to hit even the current inflation targets, which is why the above imaginary stimulus would necessitate fiscal action as well. But equally, other factors could exacerbate the inevitable bond market sell-off. Higher inflation tends to be more variable and should hence command a higher inflation risk premium to be built into bond yields. Also, longer-term real rates should move up, reflecting the reduced probability of the economy getting stuck at the effective lower bound for nominal interest rates in the future. Such a major change of tack would also put in doubt central bank credibility, inviting a further risk premium to be built into yields. In short, it would be difficult to orchestrate such a policy shift without a significant shock, given the way inflation expectations have been anchored in the developed world over the last generation. Nevertheless the mind game makes you wonder: why is two per cent the typical inflation target in developed economies anyway? The destructive effects of high (and variable) inflation on the economy are well known, so why not target zero? One reason is measurement bias. Substitution towards cheaper items and also the effect of quality changes on prices lead to an upward bias Michal Jezek Konzept 16 Konzept 16 of measured inflation relative to one actually experienced by consumers. For example, a study commissioned by the US Congress in 1996 found that the CPI overstated annual inflation by 1.1 percentage points. While some methodological changes to consumer price indices have been made in an attempt to correct for some elements of the bias, its existence played a role when current inflation targets were formulated in the past. Moreover, the size of this bias is quite uncertain and is likely to vary over time. Thus, policymakers prefer to err on the side of caution as they view slipping into deflation a bigger potential threat than small inflation. The reason is they know how to control the latter with standard interest rate tools given that small inflation, as opposed to deflation, makes hitting the nominal interest rate floor less likely. Also, given the social aversion towards (nominal) wage cuts, small inflation makes it easier to reduce real wage costs during recessions, thus dampening the impact on employment. These arguments have led to a preference for small but positive inflation. However, there is no clear reason why two per cent is the right answer, except that it could be viewed as “close enough to zero”. So why not revise it higher? The primary benefit of an upward revision to inflation targets has been stated already: easing the constraint on monetary policy that arises from the lower bound on nominal interest rates. While the central bank can embark on quantitative easing when rates hit the floor, such unconventional policies are of limited effectiveness when compared with straightforward rate cuts. This benefit could be especially large if the recently revived secular stagnation 3 hypothesis turns out to be relevant, which would entail a higher frequency of the economy being stuck at the lower bound than was the case historically. What would be the costs of higher inflation? As mentioned above, higher inflation tends to be more volatile, with less stable inflation expectations. The increased Starting from scratch, a four per cent inflation target might be superior. However, the transition costs of a switch make such a reset undesirable. Konzept 17 the current two per cent targets. However, the transition costs of a switch make a reset undesirable. There are arguments for an unintended (wink) short-term overshooting of the inflation target in order to pump oxygen into a deeply depressed economy struggling under debt overhang. For now though, sticking to current targets seems to be the winner of the least-ugly-policy contest. This problem may even go away naturally in the future if electronic money replaces (anonymous) paper currency. Then, the central bank will be able to set rates arbitrarily below zero. Freedom from the current constraint on nominal interest rates would be felt throughout the population of savers earning minus five per cent in order for their economy to escape a major recession. However appealing central banking with a higher inflation target may be, we cannot start the world from scratch. Therefore, the decisions of everyone in the economy, from businesses right down to children with pocket money, will continue to revolve around the established two per cent mark. 1 See, for example, Ball, L. (2014), The Case for a Long-Run Inflation Target of Four Percent, IMF Working Paper. 2 Blanchard, O., Dell’Ariccia, G., Mauro, P. (2010), Rethinking Macroeconomic Policy, IMF Staff Position Note. 3 For a review, see Teulings, C., Baldwin, R. (eds.) (2014), Secular Stagnation: Facts, Causes and Cures, VoxEU.org Book. uncertainty would have a negative impact on credit intermediation, reducing long-term investment. Perhaps surprisingly, some economists have also invoked the so-called Friedman rule which states that it is optimal to target deflation equal to real interest rates, and so increasing the inflation target would make matters worse. The reason? Money is a social good that is essentially costless to provide, yet the cost to its holders equals the foregone nominal interest rate. To close the gap, nominal interest rates should be zero so that the real return on both money and default-free bonds is the same. This argument does point out a monetary inefficiency (and indeed contributed to the Fed’s authorisation by the US Congress to pay interest on excess reserves of commercial banks since 2008). But it ignores a much bigger inefficiency. It is unlikely that the actual (opportunity) costs of holding cash are comparable to the macroeconomic costs of being stuck at the effective lower bound, wasting the economy’s potential. After all, there is a reason why the latter is aptly called the “liquidity trap”. More important is that a hike in the inflation target would represent a breach of contract between the authorities and both households and firms that have organised their economic relationships on the basis of that. It would not be just a temporary crisis- fighting tool to help the economy deleverage. It would be a permanent shift in policy ostensibly based on a reassessment of the likelihood of interest rates hitting the floor in the future, with no guarantee it would not be reassessed again at some point. Policy credibility would take a heavy hit, not least because creditors staring at their losses would feel the government just took a convenient route to erase part of its real debt through a surprise inflation tax. Why should policymakers be trusted ever again? Starting the world from scratch, a four per cent target might, from the macro risk management perspective, be superior to Konzept 18 No one really wants to talk about it, but there is a humongous contradiction at the heart of the ongoing share buyback mania. John Tierney Share buybacks— truths, myths and technical bits 19 Konzept On the one hand companies and investors are completely in love with them. For example, over 400 companies in the S&P 500 bought back shares last year, spending $575bn in the process. That is about three per cent of the index’s market cap and is up 14 per cent from 2013, nearly matching the $600bn record set in 2007. What is more, the Buyback Index (a subset of 100 companies with the highest buyback ratio) has outperformed the S&P 500 by almost a third since 2010, as well as over most periods since 2000. 1 On the other hand anyone who is anyone is quick to excoriate buybacks. The Wall Street Journal, the Financial Times, the Economist, academics and bloggers galore regularly attack buybacks for all kinds of reasons. 2 Among the common themes: buybacks either do not add shareholder value or they destroy it; companies are hurting themselves and the economy by not using the cash to invest; companies buy high rather than low and are hence terrible investors; buybacks enable executives to manipulate earnings and compensation. More confusing yet, there is also a view that lies somewhere in between these two extremes. Agnostics believe that buybacks are a more flexible and tax-efficient way for companies to return excess cash to investors than regular or special dividends; and that they allow companies to offset dilution from stock or stock option grants to executives or to take advantage of share prices below “intrinsic” value. The trouble with trying to assess these disconnected positions is that elements of truth as well as complete tosh are put forward by both sides of the debate. For example it is ridiculous to suggest that buybacks make a firm more valuable. Likewise, it is completely wrong to argue they have led to less investment – a common gripe in the press. The reality is that it is tricky subject and understanding stock buybacks is as much an exercise in behavioural finance as in corporate finance theory. But that does not mean we should not try. First some history. Stock repurchases only came into vogue in America in 1983 when the Securities and Exchange Commission issued Rule 10b-18. This provided the safe harbour for share buybacks. Previously there was no rule against buybacks per se, but there was a risk they could be considered market manipulation under anti-fraud provisions of the Securities Exchange Act of 1934. Buybacks for S&P 500 companies soon climbed from about two per cent of operating cash flows to over 10 per cent, while dividends remained steady at about a fifth. Buybacks offered a significant tax advantage as dividends were taxed as ordinary income while gains on shares were taxed at capital gains rates, and could be deferred for investors who did not sell their stock. In 1993 a new variable came into play when Congress imposed a cap of $1m on senior executive compensation that could be deducted for tax purposes. But the law conveniently left a loophole by exempting performance-based pay from this cap. As a result, Konzept 20 21 Share buybacks—truths, myths and technical bits average stock-based compensation rapidly rose from about one-third of total pay to two-thirds in 2000 and three-quarters in 2013. In parallel repurchases rose steadily to about a quarter of operating cash flow by 2000. Then, in 2002, qualified dividends (on stock held roughly 60 days) were taxed at the capital gains rate instead of as ordinary income. This should have reduced the allure of buybacks for investors, but instead buybacks soared to half of operating cash flows by 2007. After a steep fall during the financial crisis, buybacks rose steadily again and consumed 35 per cent of operating free cash flow last year. The underlying idea behind SEC Rule 10b-18 was to provide a more flexible way for firms to manage their capital structure and shareholder distributions. As the rule now stands, a company must announce the buyback plan and its size, but there is no obligation to follow through. A maximum of 25 per cent of the average daily volume over the past 30 days can be purchased on any given day. There are, however, restrictions on buying at the beginning and end of the trading day as well as during deals or when securities are being issued. Companies must also disclose repurchase activity quarterly, usually as part of quarterly reporting process. Even so, the process is wanting in transparency, and hence, more friendly to companies than shareholders. As with many of the deregulatory actions and financial innovations dating back to the 1980s what started as a good thing may have morphed into too much of a good thing (securitisation might be another example). While buybacks have benefitted equity investors and given companies an outlet to distribute excess cash, the explosion in repurchases over the past decade cannot only be explained by rational financial management. As critics suggest, there does seem to be an amorphous but real link with executive compensation since buybacks superficially goose returns and therefore pay. (See box at the end of this article for an example of how buybacks affect earnings per share and price/earnings ratios). Tax systems usually provide yet another incentive to favour buybacks. Dividends are taxable to investors who receive them whereas share buybacks are taxable as capital gains only to investors who sell their shares; remaining investors who elect to hold defer taxes until they sell their stock. The deferral option potentially benefits most investors given that buybacks usually amount to no more than a few percentage points of market capitalisation in a given year. Furthermore, if debt is used to finance share buybacks (or dividends for that matter), the resulting interest expense is tax-deductible, which may reduce the firm’s weighted average cost of capital. Generally, however, this shouldn’t move the needle much from a valuation standpoint unless the buyback is very large. Information asymmetries between investors and management teams are another potent issue to consider when evaluating buybacks. Changes in dividend policy and share buyback plans are often viewed as signalling devices. If a company raises its regular dividend this may suggest a growth trend and cash flow generation ability sufficient to support the higher dividend. Alternatively, if a share buyback program or a special dividend is announced this may signal that excess cash won’t be wasted on destructive deals, capex, or expenses. This could alter valuations more than theory or tax policy would predict by reassuring investors that management is acting in their interests (alleviating what text-books call the agency problem). This is no theoretical concern. In 2011, for example, Hewlett Packard’s operating cash flows were $13bn with dividend pay outs of $840m and share buybacks worth $10bn. It also bought Autonomy for $11bn in stock but post-due diligence most of the purchase price was written off. Subsequent buybacks dropped to $1.5-2bn. It was not that HP had not paid out a lot of cash – it just did so to Autonomy’s shareholders rather than its own. RadioShack and Blackberry provide another perspective. RadioShack came under severe criticism as it headed for bankruptcy having spent some $1.5bn on buybacks between 2005 and 2011 – worth roughly three-quarters of its total assets in 2005. The maker of the iconic Blackberry device, was similarly criticised for spending $3bn between 2010 and 2012 – roughly 30 per cent of its assets. Perhaps both could have soldiered on a bit longer had they instead paid wages or invested the money. But they were doomed and better served giving what was left to shareholders. The point is that investors often cheer buyback announcements by bidding up stock prices because they are suspicious of what companies might do with excess cash. And make no mistake – excess cash is a problem. Since the dot.com bust corporate America has worked relentlessly to cut costs and improve margins across most sectors. At the height of the boom, both in 1999 and 2007, the operating cashflow margin for S&P 500 non-financial companies was about 13 per cent. Last year it was even higher at 15 per cent. Companies have further increased free cash flow by refinancing debt at significantly lower interest rates and reducing capex. But that is not to say that companies are scrimping on capex and other investment so they can splurge on share buybacks. For example, US non-residential investment (a proxy for corporate capex) is running at 12.8 per cent of output in nominal terms – below 13.5 per cent in 2007 and 14 per cent-plus during the mid 1980s and late 1990s. But in real terms, non-residential investment is at 13.3 per cent, matching the 2007 peak and well above the highest levels of the 1980s and 1990s. Over the past 15 years, companies have shifted away from investing in structures to investing in high tech equipment, software and various kinds of intellectual property. Further, most of these latter categories have experienced far less price inflation or outright deflation relative to the general price level Konzept 22 23 Share buybacks—truths, myths and technical bits while price indices for structures are some three times higher than the personal consumption price index since 2000. Drilling down into the financial data for S&P 500 non- financial and non-energy companies, it is true that capex as a proportion of operating cash flow dropped from around 60 per cent in the 1990s to below 40 per cent since the crisis. But more than half of this ostensible decline can be attributed to a change in the sector composition of the index as the share of technology and healthcare companies doubled to 40 per cent since the mid-1990s. These sectors are superficially light on capex spend because US accounting rules stipulate much of their investment activity is expensed rather than capitalised. The rest of the capex decline for the S&P 500 is due to disinflation and changes in the composition of capex (accounting data is expressed in nominal rather than real terms). 3 Home Depot is a good example of a company that drastically shifted its capex strategy in response to market changes. Before the crisis, the company spent about 55-60 per cent of its operating cash flow on capex as new superstores were built. But with the growth of internet-related retail activity, Home Depot no longer needs more physical stores. The resulting shift in focus to its online capabilities has cut capex to about a fifth of operating cash flow. The company has used the freed-up cash to resume a large share buyback program. What if it had kept the cash instead? Would critics of buybacks be happy knowing the money might get ploughed into more and potentially redundant super stores? Or would they prefer management diversify within the big-box retail sector, perhaps by buying Sears; or how about turning that excess retail space into cloud computer server farms? As in the example above, it is true that if the cash is returned to shareholders it cannot be reinvested in the business. But that is a problem only if management is able to invest in projects where expected returns exceed the cost of capital. Otherwise, the general rule is that shareholders are best positioned to deploy the cash into new and more productive investments themselves. Unfortunately there is a more questionable aspect to Home Depot’s buyback strategy, which leads us to another aspect of share buybacks worth considering. Over the past six quarters the company’s buybacks climbed from about half of operating cash flow to 100 per cent while capex and dividends have remained at about 50 per cent. In other words one-half of its buyback programme over the past year was financed with debt. Funding buybacks with debt has ballooned in popularity in recent years. Looking at the 329 non-financial S&P 500 companies that repurchased stock in the past year, net cash flow (operating cash flow less capex and dividends) covered 93 per cent of repurchases, implying the remaining 7 per cent or about $40bn was financed with debt or other funds. A more telling statistic is One-half of repurchasing companies had insufficient net cash flow to cover their share repurchases. For these companies about 60 per cent or $175bn was presumably financed with debt. Konzept 24 Konzept 24 There is scant indication that companies are thinking like investors and trying to buy low and sell high. Buybacks at S&P 500 companies peaked in 2007 at $580bn, plunged to $133bn in 2009 and have since recovered steadily with a soaring market. Konzept 25 25 Share buybacks—truths, myths and technical bits that about one-half of repurchasing companies had insufficient net cash flow to cover their share repurchases. For this subset about 60 per cent or $175bn was presumably financed with debt. It is hardly comforting that the last time such a high proportion of the cost of buybacks was debt funded was the first quarter of 2008. Back then 60 per cent of companies had insufficient net cash flow to cover repurchases. It is also worth considering that the actual amount of repurchases financed with debt is probably higher than these figures suggest as many tech companies that have positive net cash flow actually issue debt to pay for buybacks rather than repatriate and be taxed on cash flow generated abroad. There is no straightforward way to identify domestic and foreign cash flow in standard financial statements. With so many companies using debt in lieu of cash flow, it is obvious that the buyback phenomenon is no longer about using up excess funds. In essence companies such as Home Depot are using the current low interest rate environment to borrow cheap debt and lever up – buying their stock on what amounts to margin. But why would they do this? Apart from trying to reduce the weighted average cost of capital, the most often-cited rationale is that companies think their stock is trading below intrinsic value. Among the many problems with this argument is that in reality share repurchases are strongly pro-cyclical. There is scant indication that companies are thinking like investors and trying to buy low and sell high (or at least avoiding buying high). Buybacks at S&P 500 companies peaked in 2007 at $580bn, plunged to $133bn in 2009 and have since recovered steadily with a soaring market and improving economy. So clearly something else is going on. Indeed there is a rational explanation: senior executive compensation is primarily performance-based. Based on a survey of half the companies in the S&P 1500 index, less than a third of executive pay is made up of salary and pension benefits; the rest is equity-based short-term and long-term incentive plans covering the next year and (typically) three years, respectively. The key variable is how performance is measured. About 55 per cent of companies measure performance based on a combination of total shareholder return as well as financial and operating results (for example, earnings, revenue, margins or return on equity or assets); 13 per cent rely on total shareholder return only; and the rest (about 30 per cent) focus on financial/ operating measures. Regarding financial metrics about one-half of companies focus on earnings, split half and half between overall earnings and earnings per share. 4 Another detailed study found that earnings per share was a factor in chief executive compensation at half the S&P 1500 companies, and suggested that smaller companies tend to rely more on earnings per share. In other words a significant portion of senior executive pay can potentially be manoeuvred using share buybacks. Konzept 26 A performance-based compensation package is essentially a call option on the firm’s future performance. The chief executive foregoes a larger salary-only package in exchange for performance based pay with the foregone salary amounting to an option premium. If the firm is doing well and performance is measured by total returns or earnings per share, a boss will be motivated to increase buybacks to hit performance thresholds. Likewise during a recession or down market there is little chance of reaching performance thresholds even with buybacks. That leaves the compensation package out-of-the-money and the chief executive motivated to conserve buyback resources even if the stock appears to be trading below intrinsic value. The problem is that these kinds of compensation packages are only nominally aligned with shareholder interests. Yes, they may help boost share prices in a rising market. But some of the resources that should be transferred to shareholders end up going to executives who “earned” them through a sleight of hand rather than genuine economic performance. The voluminous academic literature on buybacks includes a number of studies showing links between buybacks and bonus arrangements. Another motivation often cited is that buybacks are used to hit analysts’ (or the company’s own) earnings per share forecasts, with management striving to avoid an earnings miss and consequent hit to the stock price. Whatever their inherent merits or drawbacks, buybacks are likely to keep increasing. Even though the dollar volume of repurchases is near 2007 levels, as a proportion of operating cash flow it remains well below this peak. If spending on buybacks was to rise from today’s one-third of operating cash flow to 40 or 50 per cent, buyback volumes would rise by $100bn and $250bn respectively, if the recovery remains on track. At this point in the business cycle there is good reason to think that companies are both finding it more difficult to identify good investment projects and executives are stretching to meet performance targets. That may be good news for shareholders. But the bad news is that further increases in buybacks will almost surely attract more attention, much of it in the vein of antagonistic criticism and possible regulatory scrutiny. Some of this will focus on certain shortcomings in the buyback model. For example, buybacks provide a high degree of flexibility but they also are less than transparent. Taxes aside, special dividends provide much the same flexibility in returning excess cash, and are highly transparent. Regulators could tilt toward shareholders, and move to tighten disclosure rules or compel companies to follow through on buyback announcements, eliminating much of the current flexibility. Another problematic issue is using debt to finance buybacks. While this may be defensible from a capital structure standpoint, it also effectively adds leverage to the stock market, potentially amplifying both the boom and bust parts of the 27 Share buybacks—truths, myths and technical bits cycle. To put it in perspective, current total margin debt is about $500bn or roughly two per cent of the market capitalisation of Russell 3000 companies. Thus, the $150bn of debt potentially being used to fund buybacks is hardly trivial. At some point regulators worried about systemic risk could look to limit the practice in ways that make the buyback strategy less attractive. But the most probable flashpoint in the coming years may be the link between buybacks and executive pay. In an environment where wages for most people have stagnated for more than a generation any device that appears to help executive pay climb steadily (by 12 per cent in 2014 and at a compounded growth rate of 8 per cent since 1992) is going to be suspect. As America moves into an election year and with income inequality a hot topic, attacks on pay practices are only likely to grow. 5 To conclude, there is no tractable way to sort out the extent to which share buybacks are used primarily for constructive purposes, such as returning excess cash, versus serving as a tool to enhance executive pay. This is not a one-size-fits-all issue – rather the answers vary by company. The problem, unfortunately, is that misinformation abounds and rather than participate constructively in the debate most companies have instead piled on the buybacks while they can. Boards would help their cause by eliminating any link between buyback activity and executive compensation. The debate over inequality will rage on, but at least companies can offset the negative perception of buybacks to ensure the tool remains in place. History, alas, strongly suggests that common sense will not prevail. The question then is: when will something go so wrong that some combination of legislative action and regulatory fiat turns what has become too much of a good thing into what is rather less of a good thing? 1 The S&P 500 Buyback Index consists of 100 companies with the highest buyback ratio (cash spent on buybacks/market capitalisation). The Bloomberg ticker is SPBUYUP <INDEX>. 2 To cite a few examples, “US share buybacks loot the future” and “Share buybacks: who really benefits”, Financial Times, April 27, 2013; “The Repurchase Revolution”, The Economist, September 13, 2014; “The Downside to Stock Buybacks?”, The Wall Street Journal, October 25, 2015 3 See US Capex don’t be depressed in Konzept #1 for a more in-depth discussion of US capex trends. It is available via the archives at http://www.dbresearch.com/konzept_landing_en/ 4 For more information about executive compensation see Performance Metrics and Their Link to Value, Farient Advisors, 2014, http://www.farient.com/performance-metrics-whitepaper/ ; 2014 Corporate Governance and Incentive Design Survey, Meridian Compensation Partners, LLC, September 2014, http://www.meridiancp.com/insights/meridian-2014-governance-and-design-survey/ ; Cheng Y., Harford J., Zhang T, Bonus-Driven repurchases, Journal of Financial and Quantitative Analysis, forthcoming. 5 See Musings on Markets: Is your CEO worth his (her) pay? The Pricing and Valuing of Top Managers, http://aswathdamodaran.blogspot.com/2015/04/is-your-ceo-worth-his-her-pay-pricing.html ; Carola Frydman and Dirk Jenter, CEO Compensation, http://dspace.mit.edu/handle/1721.1/65955#files-area Konzept 28 Let us assume a simple world of no taxes or information asymmetries. A hypothetical company has total assets of $1,000, of which $100 is cash. It is financed with $1,000 of equity and hence its market capitalisation equals its book value. There are 100 shares outstanding, at $10 per share. The company’s earnings are $100, giving earnings per share of $1 and a price-earnings ratio of ten times. Next management decides to use cash to repurchase and cancel ten shares for $100. Total assets and equity are now $900 each and there are now 90 shares outstanding at $10 each. Earnings are unchanged at $100. Earnings per share rises to $1.11 and the price-earnings ratio drops to nine. If the company had instead paid the $100 as a dividend, assets and equity would have dropped to $900. Outstanding shares would have remained at 100 and earnings per share would be unchanged. But the ex-dividend stock price would be $9, resulting in a price-earnings ratio of 9 times ($9 share price divided by $1 earnings per share). In both scenarios the value of the firm is the same whether expressed in market capitalisation terms ($900) or as a price- earnings ratio (9). In the buyback scenario the cash is distributed only to those who sell their stock while the remaining shareholders own a proportionately larger share of the company as reflected in the higher earnings per share. In the dividend scenario all shareholders participate in the distribution and their ownership share is unchanged – but the stock price falls. At the level of the firm, then, there is no difference between repurchasing stock and paying a dividend. The primary decision criterion is shareholder preference. This is in line with the famous Modigliani-Miller theorem on capital structure that the value of a firm is not affected by its payout policy. Buybacks: The optical illusion of higher returns 29 Share buybacks—truths, myths and technical bits Credit freak show—money for nothing Konzept 30 “Nominal interest rates cannot be negative.” This familiar statement from textbooks, academic papers and policy makers has shaped the thinking of economists and financial professionals for generations. (We shall skip the arbitrage proof.) Of course, laymen did not need to be told; they simply applied common sense to arrive at the same conclusion. No longer. Welcome to the topsy-turvy world in which the time value of money has turned negative – where you pay for the privilege of lending and are paid for the kindness of borrowing. Michal Jezek, Jean-Paul Calamaro Konzept 31 From unsecured corporate bonds to mortgages, interest rates do indeed fall below zero. In the past, much effort has been devoted to ensuring that interest rates in financial models cannot go (‘unrealistically’) negative. Today, valuation models have to be re-adjusted and technology professionals at banks are fixing their systems to make sure that zero is not a constraint on interest rates. Indeed, not so long ago, those of us frequently working with large bond datasets would have excluded negative-yielding bonds from samples as erroneous data, just like assets showing negative prices. Now we know better. In macro speak, the “zero lower bound” concept has to be relabelled. To be fair we should say that one part of finance where a negative interest rate has not been unusual historically is the special repo market. (A special is an asset for which there is exceptional specific demand relative to similar assets such that repo buyers are willing to offer cheap cash in exchange.) However, this is more akin to an effective securities borrowing fee. And frankly, the general public has never even heard of repo ‘specialness’. In fact, even when the Swedish central bank pioneered negative interest rates by setting the overnight rate on deposits from commercial banks at minus 0.25 per cent in 2009, it was seen as a mere technicality that was of interest to monetary policy nerds only – a tax on reserves amounting to little more than an administrative measure. In that respect, it was not dissimilar to Switzerland’s surcharge on non-resident deposits in the 1970s, which pushed the effective annualised interest rate to as low as minus 40 per cent in 1978 in an attempt to halt the franc’s continued appreciation. Yet, three years later Sweden (current deposit rate minus one per cent) was followed by the Danish central bank (now minus 0.75 per cent) and in 2014 by the European Central Bank and Swiss National Bank (now minus 0.2 and minus 0.75 per cent, respectively). Even national legislation in some countries has been caught out by this new and unusual monetary development. For instance the Czech National Bank has been reluctant to lower rates below 0.05 per cent (‘technical zero’) because statutory interest on arrears is linked to the central bank policy rate. That would allow for (nominal) reward for delinquent debtors and penalty for their creditors, which goes against the spirit of the legislation. According to the central bank’s governor, a change in the law is required to remove this constraint (even if one could argue on purely economic grounds that nominally crossing zero is an arbitrary determinant of the penalty’s real adequacy). Still, these examples are rather arcane. Negative interest rates really started to capture the popular imagination only when they spilled over from the interbank market to investors and depositors. Initially, some banks started to charge their institutional and corporate customers for large deposits in Swiss francs, Danish kroner, Swedish kronor or euros. Later, some took the Konzept 32 unusual step of charging individual depositors negative interest as well. For example, one Swiss private bank started to charge interest of minus 0.75 per cent on balances exceeding a hundred thousand Swiss francs. A mid-sized Danish bank announced it was “adjusting its software so that it could charge retail depositors when and if it decides to do so”, while a small local competitor bank went all the way to start charging 0.5 per cent on deposits. The largest Danish bank has pledged to shield depositors from negative interest rates for at least a year but admitted it may charge customers should such an environment persist for years. At least the Danish government confirmed that negative interest on deposits is tax deductible! Undoubtedly, the longer the negative interest rate environment persists, the greater the pressure on the banking system to pass such costs on to depositors. Banks drag their feet because they are well aware of their customers’ behavioural biases, namely the interest rate illusion. To many depositors, a given level of negative expected real rates feels more outrageous if nominal rates are negative than if they are positive but with expected inflation correspondingly higher – even if the latter case attracts more tax! Perforce, commercial banks ultimately have to stay commercial and so paying for the privilege of leaving money with the bank might yet annoy many a depositor. At some level of interest rates and size of balances, some of them would undoubtedly “go entrepreneurial” by withdrawing deposits (avoiding negative interest) and putting cash in a safety deposit box (for a fee) with the very same bank. Indeed, this was happening in Portugal at the height of the eurozone crisis, albeit for fear of redenomination rather than negative interest. Generally, financial repression is both puzzling and frustrating for ordinary savers. But some real fun can ensue when negative interest rates hit retail customers on the receiving end. For example, during the pre-crisis housing boom in Spain, a local bank issued euro denominated mortgages tied to the one-month Swiss franc Libor rate. With the latter falling below minus one per cent at one point and now hovering around minus 0.8 per cent, customers with mortgage rates set at Libor plus a spread of say 0.5 percentage point are now being paid interest on their mortgage by the bank. “I’m going to frame my bank statement,” says one happy customer from Madrid. Another happy customer in Denmark pays a rate of minus 0.0172 per cent on a three-year loan. She is so pleased to receive seven Danish kroner in (negative) interest each month that even loan fees higher than that do not seem to diminish her happiness. These are not isolated cases and it is worth asking how far this could go. The issue of negative interest rates has received lots of attention in Spain, for example, where nearly nine out of ten existing mortgages are of the floating rate variety. Borrowers firmly believe that it is the lender’s responsibility to assume the risk of the underlying floating rate and to honour the contractually 33 Credit freak show—money for nothing agreed mortgage spread. If the combination of the floating rate and mortgage spread is negative, so be it. Lenders think differently. Their view is that the spirit of the mortgage contract should prevail and that under no circumstances should the lender make interest payments to borrowers. More broadly, banks across the periphery where negative mortgage rates have been reported have turned to their respective central banks for guidance on this issue. According to the press the Bank of Spain is debating what stance to take, as is the Bank of Italy. The central bank of Portugal has ruled that banks would have to pay interest to borrowers if mortgage rates dropped below zero. Will investors also see bonds with negative coupons? We have seen them already, courtesy of none other than Warren Buffett. “The new security is believed to be the first security to carry a negative coupon. Despite the lack of precedent, [it] seemed possible in the present interest rate environment,” stated Mr Buffett on the day when the corresponding five-year US Treasury benchmark was above four per cent in 2002. The old question, “When Warren is selling, should I be buying?” springs to mind. In reality, this was less about his financial magic and more about the embedded warrants that resulted in an effective coupon of minus 0.75 per cent. Today is different and negative-coupon bonds could theoretically be issued without any sweetener by some high- grade public and private entities. After all, we now live in a world in which Ireland and Portugal have recently issued six-month T-bills at negative yields! In practice, we observe zero or small positive coupons with the possibility for the bond to be issued (and then traded) at a premium. For example, the Swiss government recently issued a ten-year bond with a 1.5 per cent coupon at the price of 116, yielding minus 0.055 per cent. To reassure investors, some issuers of floating-rate bonds (such as Danish covered bonds) have published a statement that if the reference rate plus a potential spread were to be negative, the floating interest rate would be fixed at zero. However, some issuers also announced that should negative coupon rates become standard, they would allow for them in future issues. Recently, coupons on senior tranches of Spanish collateralised SME loan obligations (issued pre-crisis) were fixed at zero even if mechanical application of coupon resets would have brought them into negative territory, given the absence of any zero-lower-bound language in the offering documentation. Generally, many floating-rate bond and loan issuers across Europe have started to put in place a zero interest rate floor, something previously considered unnecessary in most bond prospectuses and loan contracts. That is a pity since our collection of credit curiosities is incomplete without some issuer asking (anonymous) holders of its floating-rate bearer bonds to come forward and make a lender-to-borrower coupon payment. Konzept 34 This fascination with zero is perhaps arbitrary, however. Investors should consider for a moment the Austrian government bond with maturity in 47 years trading with a yield of 0.755 per cent in April 2015. Or a recently issued euro-denominated bond maturing in 2115 – the first 100-year bond in euros, yielding four per cent, brought to you by none other than Mexico. Such outlandish bond pricing, while still offering positive yields, paints a no less confusing picture than any of the negative-yield examples above. Is this all just proof of market irrationality? Not necessarily. Asset pricing is about relative risks and returns, with the ‘risk free’ rate being a key benchmark. As central banks drive the latter into the negative territory, asset valuation continues to be a ‘least ugly’ contest. Ignoring some market technicalities, it is mostly the safety of fixed income securities relative to riskier alternatives that has kept bonds at their current levels. That said, there has been a concern that reaching for yield – driven by dissatisfaction with historically low absolute expected returns – may have pushed up prices of riskier assets (or long-dated bonds) beyond their fundamental value. It is hard to predict whether froth or outright bubbles in some asset classes will pop suddenly if the economy normalises faster than expected or whether price adjustments will occur rather gradually as the economy slowly recovers. It is, however, quite safe to say that in this environment impressive past returns tend to be an advance on future returns. Negative yields are just the proverbial canary in the coal mine. 35 Credit freak show—money for nothing One share, one vote—one mess Konzept 36 It seems only natural for shareholders who supply a certain amount of capital to receive an equal say on matters that affect that capital. Such a “one share one vote” system is the gold standard for corporate governance, even if other models are tolerated. Yet, today one share one vote is under pressure, especially in Europe. The newly introduced Florange law in France automatically doubles voting rights for those owning shares for more than two years, thereby creating so-called “loyalty shares”, unless shareholders vote otherwise. Sahil Mahtani Konzept 37 In Italy, new legislation has already allowed a number of companies to introduce double voting rights. In addition, the European Commission circulated a discussion paper in 2013 proposing to link dividend payments to holding periods, although this initiative appears to have cooled somewhat. While in Hong Kong, Alibaba’s decision to list in New York has compelled regulators to review their longstanding commitment to one share one vote. Double voting rights should be seen as part of a broader constellation of differentiated voting rights, or mechanisms that separate the right to cash flows from the right to votes. These include dual-class shares with one class having more voting rights than the other (of the kind Google, 21st Century Fox, and Berkshire Hathaway have), enhanced director election rights, in which one class has the right to nominate more directors to the board than the other (as for Nike, the New York Times, or Heineken), non-voting shares (the Daily Mail and General Trust) and have a functionally similar effect to arrangements that lead to ownership by the many but control by the few (like pyramid structures, cross-shareholdings, golden shares and so on). How widespread are differentiated voting rights (DVR)? Jurisdictions range from the permissive (the US, Canada, France, and Sweden) to the prohibitive (Germany, Spain, and China) and the hybrid (allowing unlisted firms but prohibiting most listed firms from using DVR like Australia, Singapore, and the UK). The proportion of listed firms with multiple voting rights is between zero and ten per cent in the US, UK and Germany, while it is between 40 and 60 per cent of firms in the Netherlands, France, and Sweden. 1 It is worth subjecting the French example to particular scrutiny since it is not only the most recent but also the most spectacular example of moving away from the one share one vote model entirely. Indeed, the intent of the Florange law was not just to permit double voting rights – France has allowed loyalty shares since 1933– but to reverse the presumption of one share one vote altogether. This year will be the last that companies can opt out before the law comes into place in March 2016 and after that, capitalism in France may change substantially. Supporters of the move make two broad points. First, they are concerned about equity market short-termism, a topic that has gained currency since 2008. 2 The suggestion is that public markets promote flawed incentive structures for managers, while encouraging a preference for exit over voice among equity holders and thus reduces long-term stewardship. As a result, supporters say, managers reduce capital expenditure, take more in profit margins, return cash through buybacks and dividends and cut investment more than they should. Moreover, pressure from activist hedge funds, short sellers, and the risk of disruptive hostile takeovers is enervating and distracting to the regular Konzept 38 course of business; sometimes this can seem like uninformed outside interference. While accusations of short-termism in public companies can often be overstated, some studies do show higher investment rates in private firms. Moreover, following the transition to public equity markets, companies have been shown to pursue less risky innovation, in comparison with those that withdrew their IPO filing and remained private. Fans of DVR point to countries such as Sweden, where almost half of listed firms have double voting rights, and certain blockholders like Industrivärden and Investor AB have more power to appoint non-executive directors and tackle underperforming managers (although this trend seems to be waning as in recent years Swedish firms have been less likely to adopt dual-class structures than the overall pool). Over a 50 year period to 2013, Swedish equities generated the highest real return of any major market. With such tantalising hints of a capitalist utopia, loyalty shares have been proposed by those as varied as Vanguard Group founder John Bogle, McKinsey managing director Dominic Barton, and prominent commentators such as former US vice president Al Gore among others. A second reason given to support DVR structures is the rise of technology firms and the need for stock exchanges to converge on arrangements that attract the needs of fast-growing firms. This is certainly behind the debate in Hong Kong, which does not have a single listed company with a differentiated voting structure. Yet in recent years the number of mainland Chinese firms listing in America with DVR structures has been greater than those listing without them, prompting concerns in Hong Kong about whether it is missing out on a new generation of Chinese technology firms. 3 Indeed, two of the three largest Chinese internet firms, Alibaba and Baidu, are listed in New York. Venture capitalists like Andreessen Horowitz have made the argument that dual class shareholdings might be particularly useful for growth companies at certain stages of development because of the firm’s reliance on human as well as financial capital. 4 In cases where fast-growing firms have already had one or more rounds of private equity or debt financing, a DVR structure allows the company to grow while maintaining continuity of management. Excessive dilution is the alternative, but, the argument goes, it may not be in the interests of shareholders, founders, or the business because it weakens the link of the entrepreneur to the business. Moreover, controlling stakes by founders insulates them from short-term pressure to generate returns (from, say, activist hedge funds, short-sellers and hostile acquirers), and allow them to focus on the management of their business. Before delving into the more subtle arguments about combating short-termism and encouraging innovation, it is worth looking at the actual context in which the Florange law in France 39 One share, one vote—one mess has been put in place. The re-appraisal of one share one vote is occurring amidst a prevailing mood of economic protectionism and a period of budget consolidation. Ever since steel magnate Lakshmi Mittal mothballed two blast furnaces that employed 20,000 people in 2012, the debate has centred around reasserting domestic control. The prolonged economic malaise since 2010 has also encouraged foreign takeovers, something that a weak euro is unlikely to diminish. The value of announced mergers and acquisition deals by foreign bidders in France was around $100bn in 2014, more than any other top-ten world economy bar America. Economic protectionism is a predictable and opportunistic response; the parallel is with the US, where differentiated voting rights began to reappear during the takeover waves in the eighties, about half a century after they were last popular. The French government says rewarding shareholder loyalty is not a protectionist measure because loyalty is defined by time held and not by nationality. Consider, though, the precise mechanism, which requires shares to be on a shareholder’s register before they become eligible for double voting rights. Many institutional investors have pointed out the administrative complexities of the process means that cross- border investors and even domestic funds can be prevented from receiving rights to which they are entitled. In this view, the law may have high-minded aspirations but in practice privileges entrenched blockholders. Blockholders such as the state itself. The French government will be able to sell around €14bn of its €90bn of shareholdings while maintaining its current voting rights; that is equivalent to 16 per cent of the 2014 budget deficit, an attractive proposition at a time when France is under pressure to be more austere. Perhaps this explains the state’s striking interventions at individual annual general meetings this spring, behaving more like an activist than a passive investor. For instance, in April, France bought €1.2bn worth of Renault shares on a short-term basis to give it ammunition to vote down a proposal to keep one share, one vote. Plainly, it seems the Florange law helps the French state and related shareholders assert control to deter foreign takeovers and may assist it through a period of budget consolidation. Politics aside, how can we gauge the desirability of DVR structures? Empirical studies on the subject are inconclusive and suffer from inherent measurement issues like endogeneity and establishing causality (see box at the end of this article). One alternative is to imagine the trade-offs associated with a DVR structure as a pendulum between entrenched owners or contestable managers. On the one hand, large owners have an equity stake, which aligns their interest Konzept 40 with other shareholders, but they are effectively immune to hostile takeovers. Professional managers are by contrast more exposed to hostile takeovers but also have fewer financial interests. If the pendulum swings too far to favour owners, then owners may take self-serving actions that effectively shift corporate resources for private purposes. Public outrage towards these structures both in Germany and the US in the 1920s made them legally suspect for decades. In Germany, some firms had shareholders with more than 1,000 or even 10,000 votes per share, while the 1925 Dodge Brothers trial revealed the investment bank Dillon Read controlled Dodge’s entire voting power despite putting up less than two per cent of the funds. If the pendulum swings too far to favour managers, as in some cases today, then they would be insulated from the disciplining effect of the market for corporate control, and could also extract more private benefits (excessive salary, expensive perks, overstaying) at the expense of the firm. 5 Ultimately, there may be no correct position. Sometimes favouring one party will lead to a longer-term focus on the product that will boost profitability; sometimes it will lead to consumption of private benefits at the expense of minority shareholders. The outcome depends on what you do with the power. Most firms are in between – one share one vote may make sense depending on the context, and therefore must be judged at the individual firm level not at an abstract one. If ownership is mostly dispersed, as in the US where the median largest voting block for NYSE-listed firms is around five per cent, (versus eight per cent for the Nasdaq, ten per cent in the UK, 20 per cent in France and higher in many European countries 6 ) then allowing a degree of disproportionate control may be beneficial because it gives certain shareholders more incentive to monitor boards. Monitoring is an expensive and time-consuming activity that demands an expertise which many shareholders do not possess. Small shareholders can be rationally apathetic. Even large, sophisticated investors cannot often look after the day-to-day operations of thousands of companies. Consider, for instance, that the average yes vote for non- executive directors in FTSE 100 companies between 2009 and 2012 was 97.5 per cent. 7 Conversely, if ownership was more concentrated, then the presence of DVRs might enhance the dominant shareholders’ chance of extracting private benefits at the expense of minority shareholders. For the minority shareholder, the best outcome is for checks and balances to ensure that both sides are empowered to contest the authority of the other. What we can say is that it is precisely in places with many established blockholders with controlling stakes – such as France and Italy – that double voting rights are the least necessary, and are certainly not needed as a default position. 41 One share, one vote—one mess DVRs would merely exacerbate the difference in power between controlling shareholders and minority shareholders. Yet it is these places that have sought to implement them first. The Florange law’s double voting provision might have been useful to protect firms with technologically risky capital expenditures ahead of them, such as Facebook’s investment in virtual reality or Google’s in self-driving cars. But in France, they are more likely to strengthen the hand of mature firms like Orange or Bouygues that are protected by the state or a family respectively and whose performance has drifted since the 1990s. Moreover, given the increasing presence of passive index trackers, a significant chunk of the long-term shareholder base will remain relatively inert. What might have been a cure for short-termism therefore instead increases the power of insiders. With this in mind, here are three common-sense suggestions for maximising the benefits of a DVR structure and minimising the risks. First, simplicity should be a key operating principle. One serious mark against DVR structures is their complexity. Complexity is often a prelude to opacity, which reduces trust and benefits those who created the laws, often insiders. By this logic DVR structures should be permitted but not as a default system (as is being proposed in France). Where they are permitted, they ought to be transparent and minimal. A case in point is Chinese e-commerce firm JD.com’s listing, where the control provisions are numerous and unprecedented. Aside from the founder owning only around 20 per cent of equity but 80 per cent of voting power through dual class shares, the board cannot form a quorum unless the founder is present, the founder must approve the appointment of any director and the founder’s superior class shares only become regular voting shares in the event of his retirement, incapacity or death. Multiply this level of eccentricity across the 5,000 or so firms listed in the US and one can begin to see the dramatic increase in complexity and minutiae that institutional investors will potentially face. So far investors do not seem to mind – JD.com trades at just under 120 times estimated 2016 earnings – and who is to say they will not be proven right. But if shareholders ever disagree with the founder, it will be difficult to change tack without destroying significant value. Second, it may make sense to reduce the gap between two classes of shares to limit excessive concentration of power. Sweden, for instance, recently legislated that no share may carry voting rights more than ten times greater than the voting rights of any other share. It may also make sense to cap voting rights at a specified ceiling, for example until the combined voting rights of any shareholder reaches, say, Konzept 42 30 or 40 per cent of the total available votes (though it is rare that blockholders will accept lower voting rights than their economic interest). It may also make sense to put some limit on the scope and duration of these structures. For instance, some firms have restriction on transfers, that is, multiple voting right shares must convert into one share one vote shares if ownership is transferred to persons who are not affiliated with the original holders (for example, Facebook, Google, Linkedin, Zynga). Other firms have rules that incorporate a minimum equity threshold by the founders, that is if at any time the founder of the company holds less than five per cent of the multiple voting shares, shares on issue must convert into one share one vote shares (58.com, Baidu, JD.com, Weibo). Tech firms mostly restrict inheritance of dual-class shares after the founder’s death. This restriction suggests some effort to reduce the potential distance between shareholders and founders. Finally, there may be too much focus on share structure which, unless abused, is merely one means to a system in which firms can thrive and minority investors get a fair deal. One measure of minority investor protection is the control premium. This is the extra price paid by shareholders for a controlling stake, which is an indicator of the level of private benefits associated with control. Studies agree that such a premium exists, on average equal to 10 to 14 per cent of the equity value of the firm. This is because controllers can potentially enjoy 100 per cent of benefits they take out of a company whilst suffering a smaller downside (though this is offset in a reduction in the value of their equity stake in the company). Yet what is clear is that control premiums tend to vary dramatically across jurisdictions. They are much smaller in systems with better protection of investors, better tax enforcement and more media pressure (for example in Hong Kong, France, the US and the UK, it is around 1-2 per cent while in Argentina, Italy, Austria or Mexico, it is between 25-40 per cent). 8 That suggests voting is not as important as what really counts, which is the protection for minorities, and the implications for the equal treatment of shareholders. Looked at this way, Sweden generates excess returns not because it tolerates dual share-class structures, but because it is Sweden, with its culture for fair-play and a sound regulatory apparatus that imposes order on jostling capitalists. Hence, it may be that differentiated voting structures only matter at the extremes, if the environment is too permissive towards owners or towards managers. Even French capitalism will probably not be hurt by the Florange law, though it is certainly self-serving and unnecessary, because France has other unique strengths, not least its Cartesian educational culture 43 One share, one vote—one mess and technological superiority that will see it through. A company is not a natural thing. Its habits are formed by human convention, conditioned by history and habit. The idea that tweaks in a company’s by-laws could transform the behaviour of managers, blockholders and minority shareholders is implausible. There is an argument to see voting as merely the big fin you can see above the surface. 9 “The vast bulk of your engagement is actually below the surface,” says one asset manager. At best voting can stop bad things from happening but it cannot ever really make good things happen; that requires commitment and stewardship. What matters ultimately is what has always mattered: fostering an equity culture in which managers manage for inclusion. That has as much to do with how people think they ought to behave as with how the law constrains them to. 1 For Europe, Institutional Shareholder Services, Shearman & Sterling LPP, European Corporate Governance Institute, 2007, “Report on the Proportionality Principle in the European Union,” 2007. For the US, IRRC Institute and Institutional Shareholder Services, “Controlled Companies in the Standard & Poor’s 1500: A Ten Year Performance and Risk Review,” 2012. 2 Pursuing DVR out of concerns about short-termism goes hand-in-hand with proposals like securities transaction taxes, capital gains tax reforms for buy-and-hold investors, compensation plans involving restricted stock, abolishing quarterly earnings reports, and escrow accounts and clawback provisions. 3 Hong Kong Exchanges and Clearing Limited, “Weighted Voting Rights: Concept Paper,” Aug 2014 4 “A Fight in Silicon Valley: Founders Push for Control,” Wall Street Journal, Lublin, Joann S. and Ante, Spencer E., July 12 2012 5 There is some evidence for all of this. For instance, the incidence of CEO turnover is slightly lower for dual-class shares than single-class shares, consistent with the hypothesis that dual-class shares entrench incumbents. See Smart, Thirumalai, and Zutter (2008) 6 Armour, J., and Gordon, Jeffrey N., “The Berle-Means Corporation in the 21st Century,” 2009 7 Zimmerman, H., Oral Evidence for the Kay Review, Business, Innovation and Skills Committee, United Kingdom Parliament, 26 February 2013 8 Dyck, A., and Zingales, L., “Private Benefits of Control: An International Comparison,” Journal of Finance, 59, 537-600. Study looked at transfers of 393 controlling blocks of shares in 39 countries 9 “The Kay Review of UK Equity Markets and Long-Term Decision Making,” Final Report, July 2012 10 Around 3 per cent discount in Masulis, Wang and Xie, “Agency Problems at Dual-Class Companies,” Aug 2007 and 18 per cent in Smart, Thirumalai et al “What’s in a vote? The short and long run impact of dual class equity on IPO firm values”, Journal of Accounting and Economics, Mar 2008 Konzept 44 Unfortunately, empirical studies fail to reach an overall consensus. For instance, DVR studies looking at the performance of companies with dual-class share structures show some underperformance, higher chief executive pay and lower efficiency of cash resources. However, studies that examine the post-listing performance of the same company show that a dual-class structure has no impact on operational performance. Further studies that isolate the impact of existing listed companies moving to a DVR structure sometimes reveal a positive effect, and sometimes a negative effect. However, there is a broad consensus view that the presence of a DVR structure results in a share price discount ranging from three to 20 per cent. 10 Some inherent measurement problems explain this divergence of results. Take, for instance, the problem of endogeneity, that is the concern that underperformance may lead companies to impose DVR structures rather than being a consequence of them. Or to pose the question differently, do dual-class structures lead to underperformance or do underperforming managers establish dual-class structures to protect themselves from hostile takeovers? True causality is hard to establish, a significant problem in all empirical studies of DVR. A second significant problem is jurisdictional bias that prevents cross-border comparisons. Any DVR structure cannot be seen in isolation; it operates within a particular regulatory and legal context. For instance, the US may allow dual-class structures but it also allows shareholders to take private action to sue for damages easily. In Hong Kong or Singapore, greater reliance is placed on rules to prevent the abuse of control before it occurs; enforcement is primarily through regulators and class action suits are not allowed. So it is not straightforward to ascertain whether any outperformance in the US or Hong Kong was caused by a DVR structure, the jurisdiction characteristics or an interaction of the two. Finally, it is often difficult to identify the true drivers of performance even within a single company, let alone across firms. Corporate structure is just one of many variables at play. Is Nike more profitable than Adidas because it sells a better product, has a better geographical mix, or because it has a DVR structure? Similarly, the New York Times’ declining profitability over the last decade has more to do with greater competition than any particular corporate structure. Academics try to reduce this problem by using large sample sets expecting such factors to cancel out but because DVR structures are relatively rare, in practice the sample sizes are quite small. Do differentiated voting rights affect performance? One share, one vote—one mess 45 Work crisis— a divided tale of labour markets Konzept 46 Work in developed countries is arguably going through the most significant transformation in generations. For the first time since the industrial revolution new technology is destroying more jobs than it is able to remobilise. And as ever less labour is needed to produce the same output, it is becoming clear in some countries that growth is now possible without rising employment and wages. Such a profound change is bound to have immense economic and social implications. Aleksandar Kocic Konzept 47 Let’s start on the economic front. Although this change seems to be a bonanza for company profits here and now, in the long run this is a destabilising trend for the entire economy. Our economic model is placing impossible demands on the recipients of wages. They are expected to perform the impossible task of supporting higher consumption, which accounts for an ever growing fraction of output (almost 70 per cent in US), in the face of declining wage and rising living costs. Credit was thought to be the magic that bridges this imbalance, but in turn has caused an unsustainable borrowing and balance sheet crisis from which it is difficult to engineer an economic and social recovery. However, the disappearance of work in work-based societies is no longer only an economic issue. It is indeed a wider social and political problem and a crisis of the entire system of values. How have developed economies arrived here? Of course there are economic, social and political forces that need to be understood. But what lies at the heart of the problem is that industrial societies have reached a point where work is the Siamese twin of life. This has not always been the case. It used to be enough to work as much as it took to derive an income necessary for subsistence, rather than earn more by working as much as possible. Indeed, early factory bosses used to pay meagre wages precisely because people were unwilling to work long hours. Suddenly in order to survive, every hour possible had to be worked. Labour became part of a reality indistinct from everyday life. Living to work is not such a problem when labour is scarce – workers have some bargaining power. At the extreme they could refuse to work in order to win concessions. As long as margins are high, however, there is usually enough money for everyone. Problems begin when margins start to compress. Cost cutting either eliminates jobs or forces workers to accept lower wages. Or as a consequence of innovation, work ceases to be the main productive input and wages the main production cost. Output becomes more a function of capital than labour with workers losing bargaining power as they weigh up poorer conditions and unemployment. This explains the modern feedback loop of higher profits and stagnant wages. Innovations and new technologies are irresistible for capital. They are a source of rent as prices no longer reflect production costs, but contain a scarcity premium. Because of that, profit centres always compete in terms of their capacity for innovation. However, innovations also reduce the need for workers, which in turn dampens wages, which in turn boosts profits, which in turn is ploughed back into investment in new technologies that further reduces hiring. Konzept 48 This dynamic is seen clearly in the last 50 years. The post war US economy, for example, shows labour’s share of output steadily declining while consumption accounts for an increasing proportion. From the 1950s to the 1970s, wages accounted for between 62 and 66 per cent of output. In the last quarter of the century, the range moved lower to between 61 and 64 per cent. But in just a decade post the millennium the labour share fell by another six percentage points. Meanwhile the gap between the returns on capital and labour also widened. Of course, this has been partly a function of institutional changes and policies, and partly a consequence of globalisation and general trade openness as labour intensive jobs migrated to emerging markets while developed market jobs became less labour intensive. Innovation in the form of technological change and improvements in communication have also raised marginal productivity and return on capital relative to return on labour. The big problem with all of this is the sogginess in wages has occurred while the consumption contribution to output has continued to grow. Indeed, as wages fell in the decade post the new millennium, consumption was a whole five per cent higher. Sure this is not a new trend. From the 1960s to 1997, consumption rose from 59 per cent to 64 per cent of output. The share accelerated to 67 per cent just a couple of years later and has continued to grow reaching 68 per cent at present. With the wage share in decline, all this spending has only been possible because debt has exploded and household liabilities expanded. While, for most of last century, growth in the household debt ratio (liabilities divided by assets) was commensurate with consumption, it went into overdrive in the late 1990s. From the 1960s until 1997 the consumption share of output grew fifteen basis points per year – about five basis points faster than the growth in debt. But from 2000 to 2007 the debt ratio grew more than twice as fast as consumption. Effectively, this has been a transition from public to private deficit spending. Credit expansion went beyond supporting consumption. It also compensated for the increase in the basic costs of living as the role of government spending changed. Consumer credit became a surrogate for wage hikes, mortgages satisfied the supposed right to home ownership, student loans replaced free education, while public healthcare was gradually pushed out by private insurance. And hence another feedback loop negatively affects work. Low wages means a greater reliance on credit, which leads to higher living costs. Thus more people are forced into working (for example, partners) and into working longer hours, and sometimes holding more than one job. That in turn increases 49 Work crisis—a divided tale of labour markets If lost growth is the economic effect of the crisis, then unemployment is its social counterpart. Okun’s law suggests economic and social responses to crises and recoveries are coordinated. Konzept 50 the labour surplus, lowers wages and amplifies a reliance on credit which increases living costs further. So less labour produces the same output as before with fewer buyers capable of paying for it. As household balance sheets become stretched, it is hard to see how the current policy response to this crisis – that is, keeping rates low in order to foster cheap borrowing aimed at boosting consumption – is effective. If lost growth is the economic effect of the crisis, then unemployment is its social counterpart. Okun’s law suggests economic and social responses to crises and recoveries are coordinated. Deviations of output growth from potential should mirror deviations of unemployment from the rate at which inflation does not accelerate. Any persistent departure from Okun’s law raises the possibility an economic rebound could happen without social recovery and vice versa. Just such a breakdown between the economic and social recoveries seems to have happened since the financial crisis. For example, despite a steady decline in the unemployment rate, growth in output is trailing way behind. This is largely a function of changes in the labour force. While the unemployment rate has fallen towards pre-crisis levels, it remains high when accounting for workers that have to work part-time for economic reasons. The duration of unemployment also remains two to three standard deviations from pre-crisis mean. Jobs are getting filled, but under different conditions and at lower wages. It is worth asking what might be happening here, especially as job rebounds combined with minuscule wage improvements have become a signature of the economy in the 21st century. During recessions wages have declined fast with increasing unemployment, but not the reverse – or in economics jargon, the hysteresis of the Phillips curve. Wages remain low and sticky, showing only a marginal response to declines in unemployment. The answer is pretty obvious to see in the unemployment data. The initial response to crises is to reduce the labour force. Companies then want maximum flexibility in the face of the recession as well as the chance it gets worse. Therefore it makes perfect sense to replace permanent long-term employees with temporary workers to save on production costs and benefits expenses. And should there be a recovery, wages lag behind. In some respects it is as if the economy is being pulled back towards the early industrial age. Back then, as mentioned above, the unwillingness to work beyond a subsistence level of survival caused employers to pay lower wages to ensure long hours were necessary in order to earn enough to live. Demand for labour was high, but workers were reluctant to work; the opportunity of earning more was less attractive than working less. 51 Work crisis—a divided tale of labour markets Today the end result – low wages – is the same but the causality is different. Late 20th century economies grew only if people consumed beyond their needs. But, for this to happen, they have to borrow more and more, especially if their liabilities continue to grow. For that, they need jobs, but jobs do not pay. So, they have to work harder, and put in longer hours, in order to survive. Unlike in the early industrial period when the scarcity of labour was the dominant factor, in post-industrial economies the supply of labour is high and costs of living high as well. What is more, the preindustrial concept of “enough” is nowadays redefined by credit. Ease of borrowing changes what subsistence means. Seen from a worker’s perspective, the negative effect of the increased efficiency of production, brought about by technology, is being offset by credit. Credit naturally extends what our needs are, and recalibrates how much we have to earn for survival. But loans need to be paid back. That is why, despite all technological advances, there has not been a commensurate decrease in working hours. Is there any hope in repairing the relationship between economic and social recoveries? Can one happen without the other and what can be done to make wages responsive again? For the answers to these questions to be positive, change needs to occur along two axes: economic transformation (namely, fiscal policy) and the transformation of the labour force (that is, social change). Policy change would work most effectively through government spending. Over the last five years government spending in America has been decreasing not only as a fraction of output, but in absolute terms, partially in response to lower tax receipts. Given an already high deficit and unfavourable demographics, however, fiscal tightening faces strong opposition. Hence any stimulus would have to be along the lines of non- Ricardian spending. This in turn would cause non-trivial reshaping of the political landscape. Meanwhile the required transformation of labour may already be happening. With the rise of the knowledge economy and a de-emphasis of manufacturing, workers are dividing into four main categories: inventors, educators, salesmen and labourers. The first three can be thought of as over the counter jobs, in they always demand personal skills that cannot be fully automated. Labouring, however, is purely exchange work, requiring nothing unique. Such jobs are an extension of assembly line work, however in a wider context today including technical and intellectual skills. These workers are replaceable and expandable. Konzept 52 Based on current trends, exchange workers face a bleak future where only their time will matter, to be exchanged for a temporary salary. The end game is job auctions, where, say, a finite term 200 hour lot would be sold. Companies have maximum flexibility at the expense of a labour force with minimal bargaining power. Workers with superior skills could demand higher pay to help smooth their consumption across periods without a job. Intermediaries would manage stables of workers with standardised skills on whose behalf they pitched for part time jobs. For companies this increased flexibility reduces the pressure to have a long-term view and strategy. Instead they can take short-term tactical positions and quickly adjust their labour costs to different market conditions. In the extreme, the demand for labour disappears completely – everyone works for himself. This is a transformation from a society of workers to a society of employers. The ultimate irony is people employ themselves but end up working long hours and paying themselves poorly. 53 Work crisis—a divided tale of labour markets German model— has a consensus economy reached its limit? Konzept 54 The German economy is basking in the sunshine. Output growth was much stronger than anticipated in the final quarter of last year. Even though the 2015 data has been on the weaker side so far, consensus expectations for growth have climbed up to about two per cent for this year and the next. Corporate sentiment has also continued to improve over the past few months. More importantly, the steady rise in consumer confidence is finally being reflected in higher spending. All of which means that Germany’s finance minister, if he so desired, could significantly increase government spending without adding to the federal debt until 2019. Stefan Schneider, Barbara Böttcher Konzept 55 But what about all those dire warnings about over-generous social policies enacted as part of the new grand coalition government? The minimum wage, of course, was predicted to be a grave policy error by many. And lowering the retirement age to 63 was supposed to cripple the economy. Yet the results so far suggest a negligible economic impact from both these policies. According to the Federal Employment Agency, the minimum wage does not appear to have had any significant impact on employment figures. Although the number of low- paying “mini-jobs” has declined markedly, some of these may have even been converted into regular jobs that are subject to social security contributions. There has been a marked increase in such jobs in the distributive trade, for example. Given only isolated price effects from the minimum wage increase, this leads some to suggest that the new social policies may actually prove beneficial to the economy. If only a few workers lose their job because of the minimum wage, and the 250,000-plus vacancies created by those new early retirees are filled, Germany might enjoy an additional increase in income that would further boost consumption and the economy. Before getting too excited it is important to take the cost side of the economy into account. For example, if the minimum wage has not led to higher prices for the goods and services concerned, it stands to reason that corporate profit margins are being squeezed. This could have negative consequences for investment and job creation. Similarly, lowering the retirement age to 63 years should also weigh on company earnings, especially as the experienced and skilled workers tend to take up the offer of earlier retirement. There is a growing burden on pensions to consider too. And even if a new employee is successfully hired, a valuable source of company knowledge has been lost and productivity suffers. So why are these negative aspects not visible in the data so far? First, time lags are probably a factor. Second, they are currently being offset by a whole range of exceptional effects. Real income gains due to the fall in oil prices, the boost to exporters provided by the weaker euro, the ultra-expansionary policy of the ECB as well as the current immigration boom in the German labour market may be masking the potentially negative impact of misguided economic policies. Worryingly, there are growing signs that this illusory success is provoking the coalition to indulge in introducing yet more generous social policies. This poses a real risk that over time Germany could end up relinquishing the benefits of painful structural improvements made during the previous decade. The problem is these dire consequences will not become apparent until the temporary and exceptional tailwinds slow down or reverse direction. Konzept 56 Given the likelihood that this misleadingly positive economic picture could persist, necessary additional reforms may be put off for years. The consequences of delay may be non-trivial. The reform measures implemented under Chancellor Schröder’s Agenda 2010 were pretty straightforward: more flexibility in the labour market, including the creation of a low-wage segment, combined with fiscal consolidation including social security. However, while these were initiated in a more favourable global growth environment prior to the financial crisis, the economic climate today is very different. With structurally slower world trade, stronger demographic headwinds and most importantly the increasingly critical perception of liberal reforms in parts of German society and politics, key pillars of the next reform Agenda are far less clear. On the positive side, however, any decisions will be taken within an institutional framework that has proven itself over and over again. After all, Germany soundly managed reunification – a shock that pushed the system to its limits as public debt and unemployment soared. In fact the robustness and cohesion of a society can be assessed by observing unemployment and inflation as well as the development of public and private debt. These indicate whether countries are able to manage distributional conflicts or whether they are prone to time-inconsistent behaviour of its major stake-holders. Germany’s strong institutional framework stands out not only in Europe but among all developed countries. One key pillar of this framework is the Mittelstand – companies with fewer than 500 employees that account for more than 99 per cent of employment. The Mittelstand comprises many market-leading businesses in their respective sectors with strong regional ties even though they are hardly known domestically let alone outside of Germany. These businesses are often family-owned and both owners and the workforce have a strong interest in the long-term success of the firm and are prepared to sacrifice short- term gains in profits or wages for this target. Workers often enjoy life-long employment, and in-company training is common. Quite often the children of blue collar workers join the company’s management levels after finishing their university education. It is also the case that more than half of the turnover of the Mittelstand involved in the manufacturing sector is generated abroad, fostering a strong focus on global markets and international competitiveness. Such an international perspective also keeps management and workers on their toes while naturally restraining excessive wage increases. This goes a long way to explain the global success of Germany’s private sector. Along with this cooperative attitude, another shared trait is a dislike for ‘experimental’ endeavours with uncertain outcomes. This risk aversion manifests itself as a deep-rooted reservation against high private and public indebtedness, thus forming a second pillar of the stable German institutional framework. This helps explain 57 German model—has a consensus economy reached its limit? a new fiscal rule demanding a structurally balanced budget while allowing for cyclical adjustments (“Schuldenbremse”). While some flexibility for the federal budget is preserved (a structural deficit of 0.35 per cent of output), German states are required to have a balanced budget from 2020 onwards. These constitutionally anchored debt brakes are not only more strict than those in most other eurozone countries but the German rule-based mentality ensures that, for the most part, a strict application will help prevent short-sighted, unfunded expenditures. A third pillar of the German institutional framework is an inclusive social system. Studies reveal a broad consensus among Germans that the state must support inclusion. This should not be misconstrued as an unrealistic over-reach for an egalitarian society but rather the conviction that economic and political challenges can only be successfully addressed if society is not falling apart. For instance, the wealth gap in society seems to be a concern for rich and poor alike. Over 70 per cent of those belonging to the upper one-fifth of the income scale consider large social differences a problem. These better-offs tend to accept re-distribution based on the feeling that individual wealth can be better justified and enjoyed if the society as a whole is prospering. While this cooperative institutional framework has served Germany well for the past 60 years, will it withstand future challenges? By design, the system has a long response time. Major changes of political direction, observable in Anglo-Saxon political systems or some smaller countries, are less likely in Germany, given the two major “Volksparteien”, the CDU/CSU and SPD, increasingly overlap in major policy fields. Moreover, with a few exceptions the system is seen to have served the country well, achieving an economically efficient and politically acceptable distribution of income gains. Given the inert response function of the German socio-economic system, strong overarching trends are usually necessary to trigger any adjustment. What are the upcoming major challenges that can force such an adjustment? Unfavourable demographic developments that will almost certainly hurt economic growth are one. Gains for society as a whole will get much smaller and their distribution may result in outright losses for certain parts of society. With the country’s median age approaching 50 years, the share of those trying to protect their vested interests is on the rise. At the same time the declining number of new entrants into the labour market will have a stronger bargaining position. Distributional conflicts, so far uncommon in German society, could become more prominent. Worsening demographic trends will also test the relationship between the government and German corporates. The country’s comeback from the sick man of Europe in 1999 to its economic poster child a decade later owes more to the adjustment in the corporate sector, in particular its globalisation drive and strong industrial relations, than to the government’s Konzept 58 reform Agenda 2010. Now there is a risk Mittelstand companies may not think of their home-market as such a great asset, as a shrinking work force constitutes a smaller counterweight to political, environmental and regulatory burdens. In such an environment the traditional strong regional links might lose their anchor function. Besides demographics, the other factor that is worth observing is changing German attitudes towards Europe. As some of Germany’s institutional strengths are challenged in the coming years, the changes may not necessarily move the country closer to its European partners. Although more than one-third of German exports still go to eurozone members, their share has dropped by one-tenth since the crisis. In contrast, German exports to China, Asia and the US have enjoyed strong gains in recent years. Even assuming a European economic recovery, the focus of Germany’s corporate sector will probably remain outside the eurozone for the foreseeable future. While economic prospects matter for corporates, the attitude of German citizens towards Europe is influenced by the collective perception of the eurozone crisis. The increasing divergence between developments in Europe and the traditional German legalistic and rule-based habits could put increasing pressure on Germany’s relations with the rest of Europe. In recent months, issues such as the commission’s lenience with regard to budget consolidation, Greek resistance to reforms and demands for a further debt restructuring, and the ECB’s unorthodox monetary policy stand (in stark contrast to the Bundesbank’s position) all chafe against the German instinct for ordoliberalism—the idea that market-based system needs rules that provide order to function most effectively. Finally, perhaps the biggest challenge facing the current consensus in Germany is life after Chancellor Merkel. Thus far, her presence may be papering over conflicts that would otherwise gain more prominence. Germany’s economic success under the Chancellor and the unwavering belief that she will defend German interests in Europe has bred a climate of consensus among the political parties, press and the citizenry. Too much consensus may have even bred complacency. Yet Chancellor Merkel cannot stay in office forever. Only over time will it be possible to judge whether the German institutional framework she presided over can survive her passing without a major overhaul. 59 German model—has a consensus economy reached its limit? European capital markets—living up to the weight of history Konzept 60 Compared with their freewheeling cousins across the Atlantic, continental Europeans are said to be inherently averse to capital markets. While it is arguably true that an overreliance on bank funding hampers the eurozone economy, it is wrong to say attitudes are ingrained. Indeed history shows that financial market innovations have sprung out of Europe with impressive regularity. It would be unwise to dismiss yet another transformation out of hand. Bilal Hafeez Konzept 61 Of course early innovations in financial markets occurred in the Italian cities of Genoa and Venice as far back as the 14th and 15th centuries. And it was the Dutch that made some of the biggest strides in capital markets over the following two centuries. Being limited in land and possessing a weak nobility but an advanced financial centre gave the Dutch the conditions to excel. And Amsterdam had a host of institutional investors, including orphanages, poor houses, hospitals and craft guilds. Rapid population growth from 1580 to 1670 expanded the social safety net and hence the demands on these institutions. Ironically it was an orphanage, Burgerweeshuis, which played an important role in developing Dutch capital markets. Founded in 1520 it was funded by city subsidies, donations and investments. The Dutch revolt of 1568-1648 resulted in Catholic assets being seized by the Protestants, which provided additional capital for Burgerweeshuis. Initially investments were in real estate, but by the early 1700s the orphanage started to venture more into financial assets such as sovereign debt and shares in the Dutch East India Company. This set the trend for other welfare institutions to become the investor base for financial assets. The Dutch firm Jean Deutz and Soon stood out as a financial innovator. It had the exclusive rights to sell Austrian mercury to the rest of Europe. In the late 1600s, the company transformed some of its loans into a unit trust in which others could invest. This was an early example of a loan securitisation and arguably laid the early groundwork for mutual investment funds. Meanwhile, the Dutch, as well as the British, developed commodity and other exchanges due to their expansion across Asia. Unlike other colonial powers that relied on taxes and government debt to fund the growth in their empires, both Britain and the Netherlands used government sanctioned private joint- stock companies to expand in Asia – the East India Company in the case of the former (1600-1874) and the Dutch East India Company for the Dutch (1602-1799). This created a need (and demand) for equity capital to fund colonial expansion. Britain became the capital markets leader during the 18th century, but copied much of the financial innovation from the Dutch. Around this time what accelerated the development of markets across Europe was the predilection for the aristocracy to gamble. The habit was a status symbol and gambling instincts transferred easily from cards to markets. America only took the baton as lead innovator in the late 1700s to early 1800s. Its sovereign debt became more federalised under Secretary of the Treasury Hamilton (1789-1793) and the Konzept 62 Through history the baton of financial innovation has been passed from the Italian cities to the Dutch and then the British. America became the lead innovator only since the late 1700s. 63 European capital markets—living up to the weight of history banking system expanded as more state charters were given in part because states were no longer able to issue money directly themselves. Unlike in Britain, most banks issued equity to raise capital, which accelerated the development of capital markets. Also, the development of private property rights took place at an earlier stage of the creation of the state. This in turn enabled the private holding of company shares. Europeans were far from oblivious to such innovations. They were heavy investors in the US at this time, providing a large and stable investor base. Rather it was during the modern era that America really started pulling ahead of Europe in terms of developing capital markets. For example, the post-depression New Deal reforms led to a unique set-up for mortgages in the US. Government- backed entities, such as Fannie Mae and Freddie Mac, started to buy and guarantee mortgages originated by banks, and then sold them on to investors. This approach was aimed at boosting domestic housing. In addition, the US thrift crisis during the 1980s provided another impetus for securitisation. It showed how ill-equipped banks were to warehouse mortgages and other longer term assets on their balance sheets. Hence the thrust of public policy from the latter part of the 1980s until the recent financial crisis was to enable banks and capital markets to shift assets to investors in order to better manage inherent risks. Indeed, securitisation was the primary tool used by the Resolution Trust Corporation (RTC) to clean up the thrift mess, and this provided the template for convertible and private mortgage back securities, as well as other asset backed structures. Fast forward to the present. The striking difference between the eurozone and the US is that deposit-taking banks in the former originate and hold the bulk of loans and other financial assets. Indeed, 60 per cent of financial assets in the eurozone are held by banks, compared with 30 per cent in America. Thanks to much larger capital markets, US non-bank financial entities – such as pension funds, mutual funds and financing vehicles – directly provide lending and other forms of finance to domestic companies, households, and other financial institutions. Banks are not immune to the vagaries of markets, though. For one thing, most banks do not rely just on deposits to fund their balance sheets – a significant proportion is funded via wholesale markets. This introduces a market element into traditional banking. For another, the conventional view is that the bank decides the lending rate on loans, but even then, credit default swap markets may give a market price for credit on a company that a bank cannot ignore. Moreover, secondary loan markets may exist which would influence the interest rate charged on new loans. Therefore, even in seemingly bank- dominated systems, markets enter into the credit formation process in various ways. Konzept 64 This is important as a Bank of International Settlements study from 2014 showed that bank-based economies usually provide a better buffer during normal downturns. 1 They are less prone to panicked decision-making and are more likely to retain credit lines while markets shut down. However, the strong dependency on banks is a positive only during cyclical recessions, whereas it backfires during financial crises. Indeed when recessions coincide with financial crises, the cumulative impact on output is three times as severe for bank-oriented economies as it is for market-oriented ones. The aftermath of the 2008 financial crisis and subsequent sovereign crises in Europe highlight this well. During these periods, bank funding from wholesale markets became more difficult and the riskiness of the country of origin started to dominate the credit- worthiness of banks, even those with substantial operations outside their headquarter countries. In many instances, eurozone banks were forced to rely on national sources of funding or the ECB, rather than broader financial markets. Throw into the mix tougher regulations and banks have spent most their time since 2008 raising capital and repairing their balance sheets. The corporate sector has lost out on lending partly because of this. European investment, which tends to be heavily reliant on bank credit, has been weak as a result – it has only been growing at an annual pace of 0.3 per cent in the last five years compared to the three per cent pace before the financial crisis. A similar, though less acute, process has happened in the US, but importantly the economy’s credit shortfall has been picked up by other financial institutions, such as asset management companies. Indeed, assets under management have surged – they are double what they were ten years ago. Much of these assets have come in the form of bonds which extend credit to companies. In terms of the economy, the US has overtaken its 2007 peak by nine per cent in real terms, while the eurozone’s is two per cent smaller. In the labour market, the US unemployment rate has nearly halved from its 2010 peak to 5.5 per cent, while the eurozone’s still remains close to its high of 12 per cent. Therefore the weight of evidence seems to point to a greater need for capital markets to growth within Europe, particularly given bank balance sheets are still under pressure and bank funding becoming ever more national. The good news is that the European Commission has taken notice and since 2014 has started the process to have the building blocks of an integrated Capital Markets Union encompassing all EU member states in place by 2019. 2 Its proposals include the simplification and standardisation of financial products ranging from bonds to prospectuses, information sharing to make the credit assessment of smaller companies easier, encouraging investors to invest in long-term projects, SMEs and start-ups, and adapting infrastructure, law and technology to facilitate cross-border flows. 65 European capital markets—living up to the weight of history But in addition to these proposals, certain less discussed quirks of Europe compared with the US will need to be ironed out. For example, under proposed Basel III rules US banks would benefit from lower capital charges for securitisation positions than European peers. American banks are precluded by Dodd- Frank from using external ratings in determining risk weights, rather they will use an ad-hoc formula already employed by trading books known as the Simplified Supervisory Formula Approach (SSFA). For example, some estimates suggest that European banks could be subject to ten times the risk weights of a US bank for senior CLOs. All of this is laudable, but it has to be remembered that the free movement of capital was enshrined in the Treaty of Rome over fifty years ago. Yet, European capital markets remain fragmented and organised along national lines, perhaps more so than before the 2008 crisis. The one tipping point for greater capital markets union could end up being the ultra-easy monetary policy pursued by the European Central Bank. Its deposit rates have been negative since last summer, which has driven many bond markets to offer extremely low if not negative yields, especially in Germany. This should have a profound effect on the behaviour of European savers and investors to seek higher returns elsewhere. In the past, holding cash was well rewarded with real interest rates in Europe being at least one per cent higher than in the US since the 1990s. As a result, eurozone households have had a big bias to hold cash – 35 per cent of their financial assets are in cash compared with 15 per cent in the US. This has resulted in only a quarter of assets being held in equities compared with 45 per cent in America. So prolonged zero interest rate environment coupled with the Europe’s push towards capital market union could set the stage for a profound change in European markets over the next five years. 1 “Financial structure and growth”, BIS quarterly review, March 2014 2 See European Commission Green Paper, “Building a Capital Markets Union”, February 2015 and the accompanying staff working document “Initial reflections on the obstacles to the development of deep and integrated EU capital markets” References: — BIS Working paper no. 406, “Financial crises and bank funding: recent experience in the euro-area” — BIS 84th Annual report, “The financial system at a cross-roads” — Hardie and Howarth, “Market based banking and the financial crisis” — Kindleberger, “A Financial History of Western Europe” — Ed. Atack and Neal, “The Origin and Development of Financial Markets and Institutions” 66 Konzept Columns 68 Book review—the chimp paradox 69 Ideas lab—the threat of artificial intelligence 70 Conference spy—dbAccess Asia 71 Infographic—corporate America: 1995-2015 67 Konzept Book review— the chimp paradox Guy Ashton You ducked a meeting you should have attended...bailed out of delivering a tough message...felt intimidated by the crowd...sided with colleagues to badmouth a friend…took a stupid risk just to show you could…claimed credit for something you didn’t do…felt hurt and defensive in response to reasonable criticism… didn’t speak up in a meeting...broke a promise. You look back at your behaviour and are baffled. You know you are better than that. But you did it anyway. And you know you will do it again. What’s going on? Professor Steve Peters’ book offers some answers. Having worked in clinical psychiatry for over 20 years, Peters is the consultant psychiatrist to Liverpool Football Club and the England football side. He has also worked with the British Olympic cycling team and other sports stars who speak highly about how he helped improve their performance. Peters starts with his working model of the brain. He describes its seven parts, of which three make up what he calls the “psychological mind”. He then labels these three elements as the Human, the Chimp and the Computer to explain how they interact. These three brains, says Peters, are joined up, but they struggle against each other to gain control. The Human is you – rational, thoughtful, sober, disciplined, caring, focused, calm and professional. But the Chimp part of your brain, the source of feelings and emotions, developed separately in the womb and only later connected to your Human brain. The Chimp is four times stronger than the Human and can hijack your behaviour. You are not responsible for the nature of your Chimp and the Chimp is not good or bad – it is just a Chimp, after all – but you are responsible for managing it. The third part of the psychological mind is the Computer – the repository of your habits, routines and automatic responses. Both the Human and the Chimp lay down programs in the Computer. Some programs are good and helpful (Peters calls these “autopilots”), but some are destructive and destabilising. As you grow up and learn from your life, you create, adapt and strengthen the programs in the Computer. The Human is generally in charge when everything is going along quietly. Say, you are working away, dealing with people calmly. The Computer is humming along in the background with its autopilots helping you cope with environments that you are familiar with. The Chimp is asleep. But then there is a threat to something your Chimp cares about. It wakes and immediately becomes alert and anxious. What does the Chimp care about? Mainly survival, and in the jungle that is all about physical security, membership of a troop, access to food, being able to reproduce, and guarding territory. A colleague starts talking about running a project you are responsible for. Your territory is threatened. The Chimp does not like it. First, it checks the Computer for programs that deal with this sort of thing. If you are lucky, there is a nice autopilot to deal with territorial threats; it quickly calms the Chimp by telling it to take the colleague’s input as constructive. What if there is no autopilot? The Chimp gets anxious – and Chimps see things emotionally. They do not think – they react. They see the world in black and white, they jump to conclusions, and are paranoid and irrational. The Chimp’s anxiety will hijack your response – fight, flight or freeze. Peters called his book The Chimp Paradox because the Chimp can be destructive, but can also be your friend. Understanding the interplay of the Human, Chimp and Computer offers an insight into your behaviour. It can also help you build better autopilots for the Computer, and manage your Chimp. The payoff may not be an Olympic gold medal, or victory in the Tour de France, but you might become the person you want to be (the person Peters says you really are) more often. Konzept 68 Will computers one day turn against us? Once the realm of science fiction, the idea has worried some high-profile brains of late. Earlier this year Bill Gates said he was “concerned about super intelligence”. That followed a warning from Professor Stephen Hawking that humans could be “superseded”; Tesla supremo Elon Musk chimed in, saying artificial intelligence is our “biggest existential threat.” As part of the ideas lab series Peter Millican, Professor of Philosophy at Oxford University, spoke to us about AI. He started at the beginning of the 20th century with a German mathematician creating the Entscheidungsproblem (the Decision Problem). David Hilbert believed that a systematic formula could prove most mathematical problems using only a system of axioms and rules. In other words, he was looking for an effectively computable procedure that verifies if a formula is provable or not. In 1936 Alan Turing, a British computer scientist, achieved this with the Turing machine, a very simple construction of a tape and a writer or eraser to write or remove symbols. The machine is so simple it can be constructed in Lego, but it is able to solve almost any mathematical problem; it is the definitive foundation of all computers today. Turing later created his eponymous test. This is where a computer needs to fool an interrogator it is a human within five minutes of questioning. In 1966, Joesph Weizenbaum created Eliza, a computerbot that acted as a therapist. It could hold a conversation with a human by giving vague questions and answers so that its responses could be interpreted as appropriate. For many, this was the first sign that computers might possess intelligence. Since Turing, however, we now understand that information can be processed by computers in a purportive manner without bringing in consciousness. Remember that until about 1600 everything was explained by the paradigm of purpose; gods, humans and nature all had a purpose. Then beliefs shifted and the paradigm of mechanism took the stage; why do things move as they do? There are purportive objects such as humans and animals that think, have desires and move independently. Then there are inanimate objects such as rocks and water, mindlessly following external forces. Later, Charles Darwin introduced yet another paradigm to explain adaptation. Adaptation is apparently purportive, but there is no mind behind it. Suddenly it was possible to appear purportive without having an inner life or consciousness. Professor Millican stresses a common mistake made today with the link between intelligence and consciousness. There is no relationship between the two. Dogs have strong desires similar to humans, but are significantly less intelligent than us. A computer can be classified as intelligent in the way it is able to process information but this has no implications for whether it has an inner perspective. So, computers may see gains in intelligence but not develop any consciousness at all. This does not mean there are no dangers. One lies with people exploiting intelligent machines, exacerbating imbalances of human power. Consider financial trading; someone develops a machine that can beat every other trader on the market and therefore takes over the system. Alternatively, a computer could become so good at advertising or political manipulation that power falls to those with the best machines. The technology is evolving fast and can be applied everywhere. The danger therefore is that it becomes all-pervasive before we are truly ready. Professor Millican concluded by arguing that we need a social revolution to adapt to a new age of machines. Humans need to reflect on how we organise society – in particular science needs to partner more closely with philosophy to help mankind progress safely. Ideas lab—the threat of artificial intelligence Charlotte Leysen Konzept 69 For this issue of Konzept your conference spy was in Singapore at Deutsche Bank’s flagship Access Asia conference. Over three days nearly 2,000 clients met with 250 companies and listened to 70 presentations. Here is a summary of the most interesting bits for readers who could not attend. The key takeaway was a widespread sense of optimism regarding China, no doubt spurred by the equity market rally investors and companies have been enjoying of late. This positivity was personified by former World Bank chief economist Justin Lin, who said in his presentation that he expected China’s growth rate to stay at or above seven per cent until at least 2020. There was far less excitement surrounding India, however. Hence the mood overall was a mirror image of last year’s Access Asia conference where everyone was worried about China but excited about the election of Narendra Modi. Those wanting a sense of the bullishness on China could have attended any one of the presentations on e-commerce. Sector valuations may be eye-popping, but so, frankly, are the growth rates. For example, the mobile channel alone attracted 90m new online shoppers last year. Meanwhile, social media ad spend is growing at a 50 per cent annual clip and video ad spend at 30 per cent. There seems to be no stigma in China to being constantly connected; 70 per cent of Chinese say they have to be on-line at all times, double the percentage for Americans. Even though the efficacy of mobile advertising is untested, company marketing departments have decided to go for it anyway, given the incredible growth in usage. Similarly conference participants are coming round to the idea of liking China’s banks again. Although banks have rallied by about 40 per cent on a market capitalisation basis alongside the broader market, valuations suggest many more investors are yet to be convinced. The sector trades on one times book value, a price/earnings ratio of seven times and has a dividend yield of five per cent. Share prices imply non performing loans of around five per cent compared with 1.3 per cent currently. Positive medium term reforms in the financial and state owned enterprise sector provide underlying support. Nearer-term catalysts include local government finance vehicle debt swaps and the formation of an asset-backed security market. While the bullish case for banks sounded sensible enough, your spy has to admit to sniffing the occasional waft of irrational exuberance around the conference. There were the deal makers trying to justify why private, pre-IPO valuations could be higher than in comparable public markets (capital raisings reached a record $140bn across Asia in 2014, with $40bn done privately). There were Chinese property companies saying they were planning to diversify into e-commerce. Finally there was the one-half of delegates who thought a slowing China economy was the biggest threat to the region, compared with the third that worried most about the Fed withdrawing balance sheet. Indeed higher US rates were barely mentioned. In contrast, why the more muted view on India? For starters attendees no doubt felt a little chastened after being overly optimistic 12 months ago. At the same conference in 2014, almost two-thirds of delegates expected to see economic change on the ground within a year while a higher rupee was forecast by 80 per cent. This year speakers and delegates alike spoke of their worry that Narendra Modi’s window of opportunity is closing. State elections are looming (most key reforms have to be implemented at the state level) and the prime minister is unlikely to have an upper house majority for any of the next five years. What is more, while economic growth has impressed, corporate profits have lagged. Under-utilised capacity, according to one presenter, is still at a level equivalent to 50 per cent of demand. Conference spy— dbAccess Asia Stuart Kirk Konzept 70 Infographic—corporate America: 1995-2015 S&P 500 Aggregate income statement ($bn) 1995 2015E Revenues 2,660 11,324 — Energy 338 1,213 — Financials 286 1,478 Cost of goods sold 1,671 7,505 Gross margin 37% 34% Net interest expense as % of sales 2.3% 1.6% Pre-tax profit 261 1,370 Effective tax rate 36% 29% Net income 167 1,050 Net margin 6% 9% Capex 171 756 Capex-to-sales ratio 6% 7% Free cash flow 148 958 FCF margin 6% 8% Dividends 58 426 Net buybacks 30 475 71 Konzept The information and opinions in this report were prepared by Deutsche Bank AG or one of its affiliates (collectively “Deutsche Bank”). The information herein is believed to be reliable and has been obtained from public sources believed to be reliable. Deutsche Bank makes no representation as to the accuracy or completeness of such information. Deutsche Bank may engage in securities transactions, on a proprietary basis or otherwise, in a manner inconsistent with the view taken in this research report. In addition, others within Deutsche Bank, including strategists and sales staff, may take a view that is inconsistent with that taken in this research report. Opinions, estimates and projections in this report constitute the current judgement of the author as of the date of this report. They do not necessarily reflect the opinions of Deutsche Bank and are subject to change without notice. Deutsche Bank has no obligation to update, modify or amend this report or to otherwise notify a recipient thereof in the event that any opinion, forecast or estimate set forth herein, changes or subsequently becomes inaccurate. Prices and availability of financial instruments are subject to change without notice. This report is provided for informational purposes only. It is not an offer or a solicitation of an offer to buy or sell any financial instruments or to participate in any particular trading strategy. Target prices are inherently imprecise and a product of the analyst judgement. The distribution of this document and availability of these products and services in certain jurisdictions may be restricted by law. In August 2009, Deutsche Bank instituted a new policy whereby analysts may choose not to set or maintain a target price of certain issuers under coverage with a Hold rating. In particular, this will typically occur for “Hold” rated stocks having a market cap smaller than most other companies in its sector or region. We believe that such policy will allow us to make best use of our resources. Please visit our website at http://gm.db.com to determine the target price of any stock. The financial instruments discussed in this report may not be suitable for all investors and investors must make their own informed investment decisions. Stock transactions can lead to losses as a result of price fluctuations and other factors. If a financial instrument is denominated in a currency other than an investor’s currency, a change in exchange rates may adversely affect the investment. Past performance is not necessarily indicative of future results. Deutsche Bank may with respect to securities covered by this report, sell to or buy from customers on a principal basis, and consider this report in deciding to trade on a proprietary basis. Prices are current as of the end of the previous trading session unless otherwise indicated and are sourced from local exchanges via Reuters, Bloomberg and other vendors. Data is sourced from Deutsche Bank and subject companies. Derivative transactions involve numerous risks including, among others, market, counterparty default and illiquidity risk. The appropriateness or otherwise of these products for use by investors is dependent on the investors’ own circumstances including their tax position, their regulatory environment and the nature of their other assets and liabilities and as such investors should take expert legal and financial advice before entering into any transaction similar to or inspired by the contents of this publication. Trading in options involves risk and is not suitable for all investors. Prior to buying or selling an option investors must review the “Characteristics and Risks of Standardized Options,” at http:// www.theocc.com/components/docs/riskstoc.pdf If you are unable to access the website please contact Deutsche Bank AG at +1 (212) 250-7994, for a copy of this important document. The risk of loss in futures trading and options, foreign or domestic, can be substantial. As a result of the high degree of leverage obtainable in futures and options trading, losses may be incurred that are greater than the amount of funds initially deposited. Unless governing law provides otherwise, all transactions should be executed through the Deutsche Bank entity in the investor’s home jurisdiction. In the U.S. this report is approved and/or distributed by Deutsche Bank Securities Inc., a member of the NYSE, the NASD, NFA and SIPC. In Germany this report is approved and/or communicated by Deutsche Bank AG Frankfurt authorized by the BaFin. In the United Kingdom this report is approved and/or communicated by Deutsche Bank AG London, a member of the London Stock Exchange and regulated by the Financial Conduct Authority for the conduct of investment business in the UK and authorized by the BaFin. This report is distributed in Hong Kong by Deutsche Bank AG, Hong Kong Branch, in Korea by Deutsche Securities Korea Co. This report is distributed in Singapore by Deutsche Bank AG, Singapore Branch or Deutsche Securities Asia Limited, Singapore Branch (One Raffles Quay #18-00 South Tower Singapore 048583, +65 6423 8001), and recipients in Singapore of this report are to contact Deutsche Bank AG, Singapore Branch or Deutsche Securities Asia Limited, Singapore Branch in respect of any matters arising from, or in connection with, this report. Where this report is issued or promulgated in Singapore to a person who is not an accredited investor, expert investor or institutional investor (as defined in the applicable Singapore laws and regulations), Deutsche Bank AG, Singapore Branch or Deutsche Securities Asia Limited, Singapore Branch accepts legal responsibility to such person for the contents of this report. In Japan this report is approved and/or distributed by Deutsche Securities Inc. The information contained in this report does not constitute the provision of investment advice. In Australia, retail clients should obtain a copy of a Product Disclosure Statement (PDS) relating to any financial product referred to in this report and consider the PDS before making any decision about whether to acquire the product. Deutsche Bank AG Johannesburg is incorporated in the Federal Republic of Germany (Branch Register Number in South Africa: 1998/003298/10). Additional information relative to securities, other financial products or issuers discussed in this report is available upon request. This report may not be reproduced, distributed or published by any person for any purpose without Deutsche Bank’s prior written consent. Please cite source when quoting. Analyst Certification The views expressed in this report accurately reflect the personal views of the undersigned lead analysts about the subject issuer and the securities of the issuer. In addition, the undersigned lead analysts have not and will not receive any compensation for providing a specific recommendation or view in this report: Bryan Keane, Karl Keirstead, Nandan Amladi, Luke Templeman, Rineesh Bansal, Michal Jezek, Peter Garber, Nicolaus Heinen, Lea Bitter, John Tierney, Sahil Mahtani, Jean-Paul Calamaro, Bilal Hafeez, Stefan Schneider, Barbara Böttcher, Aleksandar Kocic, Charlotte Leysen, Stuart Kirk, Guy Ashton Konzept 72