"Life can only be understood backwards, but it must be lived forward."- Soren Kierkegaard
"Nature has established patterns originating in the return of events, but only for the most part." -Gottfried von Leibniz to Jacob Bernoulli, 1703
The present moment in the financial markets is a particularly difficult one for investors as Federal Reserve policy has suppressed interest rates of every duration and credit quality, creating distortions in valuation for stocks in particular sectors and perhaps in general. Many informed observers have called the recent performance of the stock market "artificial," implying that it is not fundamentally driven and that a reckoning is in order, and could happen soon. To achieve some perspective on this question I will examine some past moments in market history, considering how close they are as analogues to the present and where they differ. I will then deconstruct the present moment, suggesting the future prospects implied by the current numbers. I will discuss the investment implications, making specific suggestions as to the kind of investments to view with caution in the present environment and the type of investments which might outperform while having less risk then the market. I will provide brief discussions of a number of individual stocks as well as two index ETFs which are aligned with the times.
The Primary Analogue: The Present And The 1930s
In The History of Soviet Russia E.H. Carr argued that the greatest fear of the victorious Bolshevik leaders was the subversion of their triumph by a figure like Napoleon. The one among them who seemed the most similar to Napoleon was Trotsky, a military leader with heroic bearing, so they destroyed him. What they got was Stalin.
It's the little flaws in correlation that kill you. Revolutions, wars, market crashes, depressions, and other calamities which exist on enormous scale come with a sample size so small that no two historical moments lend themselves to perfect comparison. We are now living out the aftermath of such a moment, bearing considerable but imperfect resemblance to the 1929 Crash and ensuing Depression. The policy response this time has had some similarities and some major differences. Except for the TARP program undertaken by two Treasury Secretaries during the near collapse of the banking system (2008-9), the executive and legislative branches of government have done essentially nothing. With the almost total abdication of responsible political leadership in both branches of government, there has been no coherent fiscal response at all - less even than the 1930s. As a consequence the total burden has fallen upon the Federal Reserve.
The Federal Reserve was established in 1913 in the hopes of avoiding or mitigating events like the Panic of 1907. In the now 100 years of its existence the Fed has proved reasonably adept at smoothing the effects of the ordinary business cycle, providing liquidity and lower rates to support recoveries and withdrawing liquidity and raising rates when the economy overheated. When confronted with the sort of event for which it was created, however, the record of the Fed has been mixed. The first great test came with the 1929 Crash and the Depression of the 1930s. The Fed undertook no heroic measures to counteract the downward spiral. The initial response of political authority was similar. "Liquidate stocks, liquidate labor, liquidate farmers, liquidate real estate," Treasury Secretary Andrew Mellon famously told President Hoover, arguing that it would "purge the rottenness out of the system." Mellon narrowly avoided being impeached. Nevertheless, both President Hoover and the Fed largely stood by as the economy collapsed.
Politicians in the 1930s, including FDR, were initially more concerned with balancing the budget than with fiscal stimulus (FDR changed his view in office, but look up the 1932 campaign). The financial system was left to clear, which it largely did by 1933. The economy rallied, but incompletely. The two New Deals mitigated suffering but did not change things in a basic way. There was no clear political thinking about stimulus versus a balanced budget, or the long term versus the short. Thinking isn't what politicians do. As for the Fed, its most meaningful action was the decision to raise rates in 1937 in the face of a fragile recovery, with inflation less than 3%. The market crashed a second time, falling 50%, and the economy fell back into the Depression. As an economist whose specialty was the 1930s, Ben Bernanke understood this history thoroughly and knew he could not rely on the political leadership for much. As head of the Fed he took personal responsibility and determined, whatever else happened, not to repeat its past mistakes.
Whatever the eventual outcome, I think it is very unfair to fault Bernanke. He tried. He did reasonable things. He was creative. The trouble with monetary policy is in its inherent limitations. When historians look back on the present moment there will still be two interesting counterfactuals. We might have done a great deal more with fiscal stimulus, but it wasn't going to happen. The political leadership wasn't up to thinking it through, so we will never know. And Andrew Mellon may have actually been right, in a way. Perhaps the system needed to clear. If that is true, no mainstream historian will say it out loud for quite a long time. What the Bernanke Fed did accomplish was to save the financial system and lift the stock and bond markets. These are not small achievements. Where it has fallen short is in the failure to get traction in the real economy.
QE has had a muted effect in the real world. It has smoothed the economic path providing something that looks like steady growth but at a miserable level. In the 1930s (and in many other periods) growth fell radically, then rose radically, as much as double digits both ways, before steadying into slower growth. Nothing like that has happened this time, or seems likely in the immediate future. Unemployment has fallen very gradually amidst an involuntarily shrinking work force. This has left significant pain in the workplace and dull aggregate demand in the economy as a whole. If this is truly a recovery and expansion, it is unlike any other we have seen. The structure of interest rates is peculiar to say the least, providing an odd measuring stick for investment. The rich have done better than the poor and the middle class, not a good harbinger for future growth. This is where we stand. It has certainly created a very odd moment in the financial markets.
The Construction of the Present Moment
There are several problems with the term "normal" as applied to the financial markets. For one thing, "normal" is usually defined by "average" and can resemble the "average" of the guy with his head in the oven and his feet in the freezer. There are also problems with the way average is constructed, especially when it comes to the longer term performance of the stock market. In a paper published at about the moment of the collapse of the dot com bubble ("The Supply of Stock Market Returns," June 2001), Roger Ibbotson and Peng Chen made a good effort at calculating "normal" expected returns looking back over 75 years. Their estimate of 10-11%, reached similarly using six separate methodologies, still sticks in many heads as a reasonable expectation. Unfortunately, upon close examination it is not reasonable at all.
Publishing less than a year after Ibbotson and Chen, the late Peter Bernstein and Robert Arnott argued ("What Risk-Premium Is Normal?" Jan 2002) that the Ibbotson/Chen starting date of 1926 was cheap in terms of valuation while their end point of 2000 was the moment of most extreme market valuation in U.S. history - around 35 on one-year PE and 45 on the Shiller 10-Year Cyclically Adjusted PE (NYSEARCA:CAPE). The effect was that the valuation increase over the 75-year period added 1.25% annually - ANNUALLY! - as a component of expected market return. If this were actually true as a forward projection, the market PE in the year 2075 would have to reach more than 60! The starting PE matters. So does the ending PE.
The current moment is not any part of a secular bull market. The last three secular bulls (1921-28, 1949-55, 1982-2000) each started with a PE under ten and ended with a PE that had at least doubled. This is true whether using a single year's earnings or the Shiller 10. The annualized rate of return for these secular bulls was 18%, 15%, and 18% respectively over a significant period - 8, 17, and 18 years. A large contributor to this return was a steady upward ratcheting of PE. The starting point for the present, 2009, wasn't nearly cheap enough. It's the ending point, however, that would contain absurd figures in one or several components - either a 60-plus PE, or unprecedented profit growth, or massive inflation, or more likely all three. Massive inflation is actually the most conceivable - German hyperinflation in the early 20s was tracked by the German stock market at about a 70% ratio. But massive inflation is highly improbable at this time and would produce a very unsatisfying secular bull in any case. The present moment simply does not fit into a secular bull market. Period.
Bernstein and Arnott further argued that not only was survivor bias a factor in the returns of market indexes, but that the U.S. itself was an example of survivor bias, having been steadily rerated upward over the entirety of the period. In fact, in the 20th century the U.S. broke out from a cluster of "emerging markets" like Argentina to become the world's largest economy, most transparent market, and possessor of the world's reserve currency. This was a one-time rerating, Bernstein and Arnott argued, unrepeatable and thus like the PE component not an element of return to be projected into the future. They also noted that the risk-premium of the U.S. market (the expected return above the risk-free rate defined as yield of the long T-Bond) was a forward-looking estimate, and not the 5% average Ibbotson/Chen assumed, and in the year 2000 was actually zero - ZERO! The best case for risk premium as a forward-looking estimate at this moment is maybe somewhere around 2%, not much reward for equity risk. Bernstein and Arnott were not as brutal as they might have been about the howler in the Ibbotson paper's conclusion which argued that the extremely high PE of the year 2000 was merely a rational view of accelerating future economic growth. This is pure ivory-tower-think, and came at exactly the moment that U.S. economic growth came unstuck. Yet no part of the Bernstein/Arnott rebuttal diminished the enthusiasm of the brokerage and fund industries, for whom the Ibbotson study was an ideal sales vehicle.
That being said, the components of the Ibbotson/Chen methodology do afford a starting point for considering the construction of the present moment. Their most firmly established component of stock market return is the long-term inflation rate (3%), a number so well established that it is used as an annual increase in some liberal pensions (such as the state of Illinois). This is true notwithstanding the fact that actual inflation has run much lower in recent years and the relationship between 10-year Treasuries and TIPS implies an inflation rate just over 2% at present. The current Fed target is in fact 2%, and 85 billion a month has recently been unable to achieve it. This is a sneakily scary fact; you might reread the previous sentence. The undertow of the present moment is deflation. Inflation at even the 100-year average rate of 3% is not on the immediate horizon.
What about earnings? The number Ibbotson/Chen use for real earnings growth is 1.75%. This is probably okay, being well under average GDP growth, unless longer term estimates for real GDP growth need to be dropped from 3% to 2% as economist Robert Gordon and Paul Volcker, among others, have recently argued. And dividends? The dividend rate has actually increased from 1% in 2000 to 2% currently, but is still nowhere near the Ibbotson/Chen average of 4.28%. To get there CEOs in the aggregate would have to double their dividend payout ratio - an event for which I'm not holding my breath. The PE expansion number of 1.25% must of course be dismissed as an absurdity. We might assume that current PE will hold its own, which generously accepts the current PE as about average even though the Shiller 10 remains quite high and corporate profit margins are far above trend. Putting the components together we get a present moment with future "average" return expectations of about 5.75% nominal or 3.75% real - at best! This sort of return does not define a secular bull market. Period. The question then is: what sort of moment in the past does the present truly resemble?
Not Bull, Not Bear, Not the 30s, Not the 50s, Not Japan...Quite
The underlying economic tendencies of the past decade have many similarities to the 1930s, including the aftershock from the crackup of two investment manias, weak growth, poor creation of new jobs, deleveraging, and a pervasive tendency to succumb to a deflationary undertow. There are however some places in which the present diverges from the 30s. Politicians have been unhelpful but the Fed has been mega-accommodative. The result is something historians may call a "Contained Depression." The markets and the larger economy post-2009 have both been relatively low in volatility with a slight upward bias, nothing like the turbulent 30s. What we do not have, and we may be glad of it, is any immediate prospect of what actually brought about the end of the Depression - the massive stimulus of military employment and armament manufacture for World War II.
The era when Fed policy most closely resembled the present was from roughly 1941 to about 1957. This period is often cited in discussing end points of extreme Fed accommodation, but estimates of what may characterize the times before the Fed moves to remove accommodation tend to miss the different motives of the post-WWII Fed from those of the present policy. World War II was accompanied by price controls and the Fed suppressed rates at first to accommodate wartime spending. The immediate post-war had a large one-time pop of inflation and a brief recession, but for the next dozen years the Fed suppressed rates and encouraged mild inflation (well under the 3% average) while solid growth and currency devaluation made the tremendous war debt more manageable. Pent-up demand from wartime controls and the baby boom did the rest in producing the 1950s boom. Nothing of the sort is implied by the present situation.
The need to deleverage now exists not only at the Federal level, but at the level of households and state and local governments. The Fed may be able to essentially bury Federal obligations, but without achieving traction in the real economy it can't do nearly so much for the other forms of debt. We may well now be at about the point the economy might have reached around 1942 if we had not begun furiously building ships and warplanes. We'll never know. There is no particular reason to think that another nasty meltdown like 2007 is in prospect, mainly because we have already blown up so much of the dynamite that such a meltdown would require. The disaster itself and the Fed response have managed to dampen things down to a gray equilibrium.
This suggests comparison with the Japanese stock and real estate bubble which burst in 1989 leaving an aftermath which persists to this day. The Japanese bubble was on a scale several times that of the U.S. markets in 2000 or 2007, and the Japanese economy was even more massively leveraged. Both the BOJ and the Japanese government were very slow to take action. A truly aggressive response occurred only in recent months. Instead of a baby boom the Japanese have the worst demographics in the world. Compared with a Japanese population in actual decline, present U.S. demographics are merely dismal. What Japan did have going for it, in a way, was a culture in which employees were regarded as stakeholders, so that companies were very reluctant about layoffs. The Japanese also decided to pretend that their walking dead banks were OK. Japanese tend to save and have little household debt. The consumption part of their economy isn't particularly robust. These factors may have dragged out their "Contained Depression" but greatly mitigated the pain to individuals. The Japanese market crashed about 80%, had a few dead cat bounces, and has remained in a crashed condition up to the present year. What Japanese investors got was rather puny dividends and not much else. Thank goodness, I guess, that we're not exactly reliving the Japan experience. We are in uncharted waters.
The Investment Implications of Today's Markets
There are a few key principles:
(1) All times are "abnormal," even the ones that seem to express the fleeting moment of being statistically average. Imagine trying to figure out what was happening to you in 1935. Or 1975. Easy to know with hindsight.
(2) The investment future is always in uncharted waters. Sailors take warning: sea monsters are mainly a figment of map makers, but you never know for sure. We may actually get real deflation...or real inflation. But probably not.
(3) You can learn from the past but you don't get exact replication. Avoid rigid models and interpretations.
(4) You can also learn a lot from the composition of the present (or as Yogi Berra said, you can observe a lot just by watching), but you must be careful not to misinterpret the numbers.
(5) You should consider all long-term projections with skepticism (including this one).
(6) The biggest single problem of the present market may be the weirdness of the risk-free rate. Should we measure by the current bond rate (2 1/2%) or by the long term average (5%). Think of it this way: if we shrank every ruler by fifty per cent every American would instantly be twice as tall. In feet and inches, that is. But in absolute terms they would be the same damn height. That's the problem with all current investment returns. They are okay if measured by a tiny little ruler, but who knows when we may return to the standard ruler?
And now filtering out the mythologies and projections let's consider what the numbers actually are saying. Inflation is low, less than the long term average. This is in spite of heroic monetary actions to lift it. This suggests that it is not necessary to worry too much about inflation or own things that require inflation in order to succeed. There is another positive side to low inflation. The hurdle for returns is lower - an aspect of that tiny little ruler. Even bond yields are not quite as terrible as they look, especially if you are able to do tricks like riding down the yield curve. This is also the reason the market has bid up dividend growth stocks, too much perhaps, but not altogether irrationally. Even if inflation gradually normalizes, as it will some day, there is a good chance that dividend growth will keep up. Here it is useful to recall that in the 1950s - the point which current monetary policy seems to resemble - the inflation numbers that people got agitated about were around 3%. If inflation rises a little and rates are hiked to combat it, the valuation of stocks may experience a one time repricing downward. This is not such a terrible thing for investors with a long time horizon or who are primarily concerned with income.
Economic growth is also well below the long term average, has been so for several years, and seems likely to remain so for quite a while yet. This is also despite heroic monetary policy by the Fed in the effort to produce faster growth. Excessive debt at all levels has yet to fully clear. Heaven knows what states and municipalities will end up doing about their debts and pension shortfalls. There is no pent-up demand as aging baby boomers are about done buying stuff. Economists like Robert Gordon argue that this is the new permanent state of affairs, with real growth around 2%. Permanent is a long time, but ten years or so of continuing low growth is a good bet. Growth is therefore at a premium in the markets.
A company that can grow in this environment is a treasure almost beyond price - and that is how the market regards the Amazons, Pricelines, Facebooks, Linked-ins, Twitters, etc. Note that each of these companies and many others like them are not so much involved in the production of real goods and services as in the cannibalization of already existing production through more efficient promotion or delivery. The net effect of their cumulative existence is generally deflationary as they facilitate price shopping. But cannibalization is inherently self-limiting, and it does not create more production or employment at the beginning of the process nor more final sales at the end. It merely restructures the channels between the two and redirects the basic flow of goods, imposing new business models in place of less efficient promoters and channelers. Growth of these companies should eventually bump into a ceiling imposed by the markets they serve. What follows are a few ideas for the kind of strategies and companies that might work fairly well over the duration of these tough times.
First, think about things that wear out or get used up. Things that people or companies have to have. Things that will grow at least as fast as the economy, faster if possible. And things not too susceptible to downturns. The current market hasn't missed the advantages of sectors and companies that meet these requirements, and many of them are a bit pricey. Johnson and Johnson (NYSE:JNJ) and Abbott (NYSE:ABT) fit perfectly, part consumer, part health care, necessary products, disproportionately serving the only growing population segment (old folks), steady low-to-moderate growth, and steady product pricing, but they are pricey stocks, not wildly so, but at the high side of full value. I own them both and I'm not selling, partly because of embedded cap gains, but I'm not buying more and probably wouldn't initiate a position right now. I would look again, however, in a moderate correction. The outline of this argument applies to many stocks and market sectors.
Banks fit the story, as well as property and casualty insurance. Banks have already had their crisis. Despite some problems with the structure of interest rates well run banks, and PC companies should find the current environment reasonably friendly. Money isn't going out of fashion as a product any time soon. My favorites are Wells Fargo (NYSE:WFC), Chubb (NYSE:CB), Travelers (NYSE:TRV). Like the health care/staples, I own them, wouldn't buy now, would add on a significant correction. The same is true of a number of growthier companies - Sigma Aldrich (NASDAQ:SIAL), Amphenol (NYSE:APH), Fastenal (NASDAQ:FAST), as examples -- stocks I have owned at one time or another but find pricey today, though not outrageously so (well, maybe FAST).
Parker Hannifin (NYSE:PH) and Dover (NYSE:DOV) are somewhat slower but steady growers, have unsurpassed dividend records, and also deal in a wide range of precision engineered products which serve other businesses. Product lines of this sort provide an excellent moat. I continue to hold both but would not initiate a position at the present price. Deere (NYSE:DE), although its business is cyclical, will grow and prosper because the global production of food will grow faster than the U.S. GDP. It has a bit more debt than I like, but excellent cash flow, a dividend rate matching the 10-year Treasury, low PE, and a steady reduction of share count adding up to 20% over the past ten years. All of the above industrial companies are roughly as consistent as JNJ in earnings growth, a fact that is somewhat underappreciated by the market. All of the above are also candidates for put writes and covered call writing where the level of volatility provides enough return.
One industry which looks less pricey is energy. The integrated oils are best suited to the moment as the integrated business model allows some mitigation of fluctuations in oil prices. Cash flow at both Chevron (NYSE:CVX) and Exxon (NYSE:XOM) permits meaningful share buybacks. Intelligent buybacks which actually reduce share count should not be underestimated as drivers of EPS growth in a low-growth environment. I am less enthusiastic about the E&P companies, which are more tied to the price of crude. I prefer the oil majors because the quality of the oil industry which aligns with the current hard times is the continuing profitability of companies which extend from exploration to relatively stable final sales. Paradoxically, oil service companies may also have steady modest earnings growth as oil companies must replace reserves or find themselves out of business. My favorite is National Oilwell Varco (NYSE:NOV).
Berkshire Hathaway (NYSE:BRK.B) combines several of the business sectors and attributes which suit the times - insurance, industries which serve other industries, railroads, utilities, and staples - a rare collection of sector leaders within a corporate culture which does things the right way. Buffett has restructured exquisitely to redeploy capital internally (owning a railroad and utility is far better than owning shares). His cash position beautifully reflects the mood of the moment - invest for reliable cash flow but balance this with significant dry powder. With all of the stocks mentioned I am not so much making a specific recommendation to buy as suggesting the sort of stock favorably aligned with the characteristics of the present environment. Use them as a source for further investigation.
For the passive investor two variant index funds would be my choice for the moment - the Power Shares FTSE RAFI US 1000 ETF (NYSEARCA:PRF), a creation of the same Robert Arnott who coauthored the article on risk-premium, and the S&P 500 Equal Weight (NYSEARCA:RSP). RAFI indexes are weighted not by market cap but by fundamentals such as cash flow or book value, while the S&P 500 Equal Weight produces a similar effect by reducing the weighting of the largest market caps. Both are somewhat aligned with the long term tendency of smaller cap and value to outperform. Both these funds have a good chance of outperforming straight indexes in part by avoiding the fact that large cap size tends to chase past success. The expense ratios are around .40, more than the cheapest straight indexes but not expensive. For new investors I would recommend dollar cost averaging by investing at regular intervals and for lump sum investments of any size I would suggest dollar cost averaging so as to work up to your desired stock allocation over not less than two or three years.
Finally, a note on my personal expectation for the way this period is likely to play out. Markets and the economy will ultimately "normalize." Before the next secular bull we will probably have one more shakeout that gets the market PE down to the vicinity from which secular bulls begin. It will be painful and scary. No obvious crisis like 2007 is apparent on the horizon, but a policy mistake like the premature rate hike of 1937 or application of a kooky austerity to cut the debt or balance the budget would certainly do the trick. Meanwhile the Fed will taper and allow the Treasuries on their books to run off into oblivion. There will be no hyperinflation. The actual endgame of easy Fed policy might very well resemble the middle of the 1950s when the Fed began raising rates to address inflation which reached a terrifying 2.9%. When the Federal funds rate was increased to 3% in 1957, it proved to be the magic number. In the first quarter of 1958 GDP dropped 10% year over year, unemployment moved back over 7%, the Dow fell 20%, and people were scared out of their wits that the Depression had returned. It hadn't. The Fed promptly dropped rates below 1%, the market rallied smartly, and the economy recovered. The Fed was back in the business of tightening to curb excesses and loosening to fight ordinary recessions. Those were normal times, but who knew it.
Disclosure: I am long ABT, BRK.B, CB, CVX, DOV, JNJ, NOV, PH, TRV, WFC, XOM. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: I have cash-covered put writes on CVX, DE, JNJ, NOV,and WFC