Reviews for Andrew's books.
It has been hard to generate real enthusiasm for general equilibrium analysis in recent years. While dynamic stochastic general equilibrium (DSGE) models provide a pleasing union of macroeconomic aggregates with transparent micro-foundations, the core insights are now decades old and much of the work since has been in less consequential work of extension and calibration.
In this sense it requires either a long memory or historical imagination to picture the reaction which may have greeted Keynes' General Theory. Reading it today, one is struck both by the breadth of subjects covered and by how dramatically the style of reasoning and argumentation differs from contemporary macroeconomics. These observations are connected; it is almost impossible to imagine a contemporary orthodox approach being able to reach definitive conclusions across such a sweep of economic life.
While comparisons with Keynes might seem overly grand, it is possible to feel a similar sense of freshness at the approach taken by Andrew Smithers in his new book. He ostensibly focuses on the stock market but the treatment is wide ranging. The book sets out to challenge a number of assumptions and results in neoclassical models, with conclusions on the nature of corporate finance (listed and unlisted), financial stability, demographics and the structure of pensions, and the expense/capitalise debate in the national accounts. The conclusions are also firmly stated, and at times proudly in defiance of received wisdom. This is made possible by the style of reasoning, which leans primarily on accounting identities, observations of individual behaviour and high-level empirical evidence, rather than model solutions and calibration.
Having stated that the real return on equity is stable in the long run, Smithers provides a number of empirical approaches which broadly reinforce this. This becomes a gateway to a number of other conclusions, such as the impact of household portfolio preferences (overwhelmingly on leverage rather than equity); the solution to the equity premium puzzle (because the neoclassical representative investor is a speculator rather than a rainy-day saver); and the invalidity of the Modigliani-Miller Theorem that firm values are independent of capital structure (because leverage does not raise individual cost of equity). The alternative approach to argument is both a refreshing change and a necessary tool to reach these challenging conclusions.
Readers of Smithers' previous work will recognise a few familiar threads here. Stability of equity returns, debunking flawed market valuation techniques, and the impact of the bonus culture have all themselves been the subject of earlier publishing. But the new offering is worth the attention of past readers for two reasons. First, Smithers' thinking on these subjects has evolved in response to new data. The Anglo-Saxon bonus culture, for example, still drives underinvestment in listed firms, but may not matter as it is losing out to other more growth-oriented corporate structures. The discussion of this topic also highlights another strength of the book. Smithers' observations on how the behaviour of managers, savers and professional investors diverges from mainstream theory are, in this reader's experience, on the mark.
Second, the knitting together of these diverse threads into general equilibrium thinking is new and, to this reader, exciting. The overarching model captures some interesting features of the financial economy, but is parsimonious enough that it could in fact be absorbed into a more conventional general equilibrium model. In this sense it represents a more promising way to incorporate finance into macro than the narrow efforts to give the neoclassical economy a complex banking sector.
It is possible to take issues with the method. Those empirical observations, particularly of stability in variables such as equity returns, lifespan of fixed capital and capital-output ratios, perform critical roles in establishing or undermining theoretical relationships. In some cases, the stability can be in the eye of the beholder, especially when extensive multi-decade averages are taken. While the claim that real equity returns revolve around a mean of 6.7% is well-evidenced, it hints at fragility (what if the true rate migrates to 6.8%? 7.7%? 67%? How should we judge this?) It would be interesting to consider the potential causes and consequences were any of these mean-reverting relationships to break down in future.
In addition - another feature which this reader found in common with the General Theory - some of the accounting relationships are hard to follow in sentence format, particularly when the conclusions are counter-intuitive. Few ideas can be improved by adding rather than subtracting algebra, but there may be one or two examples in the book.
In the book's favour, though, is that conclusions are clearly stated and there has been a deliberate focus on the creation of testable predictions. Developing and executing these tests would be a logical next step if the ideas can gather the attention and debate they deserve. The other obvious departure would be to expand on the potential implications of the research for investors, both in equities and other capital markets. This could be one of the few results for a Google search of "stock market book" that does not offer investment guidance as such. But investors, along with corporate managers, regulators and setters of fiscal or monetary policy, would be well advised to read and form their own conclusions.
LONDON, April 14 (Reuters Breakingviews) - Economic theory today is far removed from what happens in the real world. Its canonical models portray the corporate sector as a single representative firm that acts in the interests of its owners. Anyone who has worked in finance knows these models are contrived. In his latest book, "The Economics of the Stock Market", veteran economist Andrew Smithers lifts the corporate veil to reveal a world in which the managers of public companies put their own interests first and seek to maximise current share prices rather fundamental values. In the United States, their actions have produced an overvalued stock market, excessive corporate debt and inadequate levels of investment.
Smithers, who started work in the City of London six decades ago and once ran the fund management arm of the merchant bank S. G. Warburg, belongs to a venerable tradition of economists whose theory is shaped by practical experience. David Ricardo started his career as a stockbroker, while John Maynard Keynes was the bursar of his Cambridge college and chairman of a life insurance company. Economic models, says Smithers, should fit with known human behaviour and be tested by data from the real world.
Theory suggests that company managers have the same interests as shareholders. In reality they have different priorities. Corporate executives aim to hang on to their jobs and enhance the value of their stock-based compensation.
If managements aimed to maximise the net worth of their businesses, they would issue shares when the cost of equity is low (and shares are highly valued in the market) and use the capital for investment. They don't act in this manner because the immediate effect of new investment is to lower a company's earnings per share. Along with issuing new shares, this tends to temporarily depress stock prices.
Instead, managers prefer to take on debt to buy back shares at inflated prices. Finance theory suggests that a company's valuation should not change whether it is financed with equity or debt. In reality, debt-financed buybacks serve to boost share prices, says Smithers. He also observes that companies seek to maintain a stable ratio of interest payments to profits. Thus, as long-term interest rates have declined, U.S. companies have taken on more and more debt to repurchase their shares.
As a result, the valuation of the U.S. stock market has significantly diverged from its fair value, says Smithers. Finance theory denies that we can identify a stock market bubble in real time: future share price movements are unpredictable. This is true in the near term, says Smithers. Over longer periods, however, the behaviour of the stock market has been anything but random. In the past 200 years, U.S. equities have delivered an annual average real return of 6.7%. Periods of above-average returns have been followed by sub-par returns, and vice versa.
This shows the stock market is governed by the principle that returns will revert to their long-run mean. Smithers suggests the best way to value equities is to compare their market price to the cost of replacing underlying corporate assets. This measure, known as Tobin's Q, is named after the Nobel laureate economist, James Tobin. The snag is that the process of mean reversion can take decades, well beyond the time horizon of most investors.
Because Tobin's Q is not a practical valuation tool, most investors prefer to compare earnings yields – a company's earnings per share divided by its share price – with bond yields. In recent years, as bond yields fell to their lowest level in history, the valuation of U.S. stocks has soared. But Smithers maintains that comparing the two makes little sense. After all, stocks are claims on real assets whereas bonds represent paper claims. Over time, the difference in their respective investment returns (known as the equity risk premium) has neither been stable nor mean-reverting.
Besides, Smithers argues, stocks should deliver a significantly higher return than bonds. Most investment is intended for retirement and savers are concerned primarily with maintaining their future spending power. Stocks are risky assets, whose value can remain depressed for long periods. After the October 1929 crash it took around a quarter of a century for the market to regain its previous peak. The marginal investor, says Smithers, requires a significant return to compensate for the market's inherent volatility.
Smithers' analysis suggests that the U.S. stock market today is perilously positioned. In recent decades, corporate managers have diverted resources from investment towards share repurchases. A prolonged period of underinvestment has put American public companies at a competitive disadvantage to foreign-owned firms. The corporate sector has also taken on near-record amounts of leverage. On a replacement-cost basis, the stock market trades at more than twice fair value. The risks of another financial crisis appear elevated, says Smithers.
Naysayers will point out that American equities have looked overvalued relative to Tobin's Q for the best part of 30 years. Besides, just because the return from stocks has been stable in the past doesn't mean that equities must deliver the same return in future. Naysayers may also suggest that the increasing importance of intangible assets has rendered Tobin's Q obsolete – though Smithers vehemently rejects this. The natural monopolies created by the internet have also allowed technology companies to earn excess returns on equity for prolonged periods.
Yet one reason why the valuation of U.S. stocks has remained elevated for so long is because the Federal Reserve has supported Wall Street with ever-lower interest rates and successive bouts of quantitative easing. Now the return of inflation has forced the Fed to reverse tack. Inflation tends to push up interest costs faster than corporate cash flows, forcing companies to deleverage and cut investment. Under those circumstances, the valuation of U.S. stocks could tumble.
Smithers was one of the few economists to warn about the internet bubble and the dangers posed by the ensuing global credit boom. His current concerns shouldn't be dismissed lightly.
(The author is a Reuters Breakingviews columnist. The opinions expressed are his own.)