John P. Hussman, Ph.D.
Last week, the financial markets mounted a striking shift back to the « risk-on » trade, as investor concerns about a recession were abandoned, and Wall Street came to believe that Europe will easily contain its banking problems. Accordingly, downside protection was largely discarded (as reflected by a plunge in the CBOE volatility index), price-volume action reflected heavy short-covering short sales, investor interest shifted strongly away from defensive sectors to speculative ones. For defensive investors, it was admittedly a difficult week, as the markets suddenly became convinced that no defense was needed, and treated defensive investments accordingly.
From my perspective, Wall Street’s « relief » about the economy, and its willingness to set aside recession concerns, is a mistake born of confusion between leading indicators and lagging ones. Leading evidence is not only clear, but on a statistical basis is essentially certain that the U.S. economy, and indeed, the global economy, faces an oncoming recession. As Lakshman Achuthan notes on the basis of ECRI’s own (and historically reliable) set of indicators, « We’ve entered a vicious cycle, and it’s too late: a recession can’t be averted. » Likewise, lagging evidence is largely clear that the economy was not yet in a recession as of, say, August or September. The error that investors are inviting here is to treat lagging indicators as if they are leading ones.
The simple fact is that the measures that we use to identify recession risk tend to operate with a lead of a few months. Those few months are often critical, in the sense that the markets can often suffer deep and abrupt losses before coincident and lagging evidence demonstrates actual economic weakness. As a result, there is sometimes a « denial » phase between the point where the leading evidence locks onto a recession track, and the point where the coincident evidence confirms it. We saw exactly that sort of pattern prior to the last recession. While the recession evidence was in by November 2007 (see Expecting A Recession ), the economy enjoyed two additional months of payroll job growth, and new claims for unemployment trended higher in a choppy and indecisive way until well into 2008. Even after Bear Stearns failed in March 2008, the market briefly staged a rally that put it within about 10% of its bull market high.
At present, the S&P 500 is again just 10% below the high it set before the recent market downturn began. In my view, the likelihood is very thin that the economy will avoid a recession, that Greece will avoid default, or that Europe will deal seamlessly with the financial strains of a banking system that is more than twice as leveraged as the U.S. banking system was before the 2008-2009 crisis.
It is certainly true that our aversion to these risks has been punished over the past couple of weeks, as investors have abandoned defensive positions in favor of speculative ones. But as always, our investment horizon remains the complete bull-bear market cycle, and there is no compelling evidence that the serious risks that we face have abated. On the valuation front, we presently estimate a 10-year prospective total return for the S&P 500 averaging just 4.8% annually (nominal), so long-term investment prospects are only weakly more encouraging than near-term ones.
While many Wall Street analysts continue to view stocks as cheap on the basis of forward operating earnings (which reflect expectations of a continued economic expansion and the maintenance of record profit margins indefinitely), the use of forward P/E multiples is a valid shorthand for discounted cash flow valuation only when profit margins reflect a level that is actually likely to be sustained over several decades. Even then, the benchmarks typically applied to forward operating earnings are actually based on historical norms for price-to-trailing net earnings.
Investors should recognize that P/E multiples are simply a crude shorthand for legitimate valuation calculations (specifically, the careful discounting of a whole stream of future deliverable streams of cash to the investor). P/E multiples subsume a whole set of assumptions regarding the entire future path of growth rates, profit margins, return on invested capital, and other factors. The common practice of valuing the stock market based on « forward operating earnings times arbitrary P/E multiple » is not only misguided – it’s an utterly disappointing display of Wall Street’s willingness to dumb-down the investment process. As investors have discovered through more than a decade of zero returns, the constant abandonment of intellectual effort comes at a cost over the long-term. This is a good opportunity for investors to review their tolerance for significant losses. My impression is that this may be the best opportunity to reduce risk that investors are likely to see for a while.
Europe: Just Getting Warmed Up
A few weeks ago, I noted that Greece was likely to be promised a small amount of relief funding, essentially to buy Europe more time to prepare its banking system for a Greek default, and observed » While it’s possible that the equity markets will mount a relief rally in the event of new funding to Greece, it will be important to recognize that handing out a bit more relief would be preparatory to a default, and that would probably be reflected in a failure of Greek yields to retreat significantly on that news. »
As of Friday, the yield on 1-year Greek debt has soared to 169%. Greece will default. Europe is buying time to reduce the fallout.
The central problem facing the global economy here is leverage. In an economy where monetary authorities are at the ready to reignite bubbles after any setback, it has been possible for banks to get, say, $10 from shareholders, get another $20 by issuing bonds, get $70 from depositors, and then go out and make $100 of investments in loans, securities, Greek debt, and other assets, hoping that by leveraging shareholder capital (« equity ») 10-to-1, they would earn a high return on that equity.
The trouble comes when some of the investments go bad. In that case, a loss of anything approaching 10% on the assets will eat through the bank’s capital (« equity cushion »), at which point, the bondholders are next in line. It can’t be repeated enough that when Wall Street talks about a bank « failure, » it means the failure of the bank to pay its own bondholders. In virtually every case, big or small, the only parties that stand to lose from a bank failure are the bank’s own stockholders and the bank’s own bondholders, both who knowingly take a risk in order to reach for return.
The only question is whether a needed restructuring is orderly or disorderly. Washington Mutual was so orderly that it was forgettable, despite being one of the largest U.S. banks prior to the financial crisis (though senior and subordinate bondholders are still fighting). Lehman was disorderly. Bank bondholders should lose when the management of that bank takes on excessive leverage, and then makes bad investments with the funds.
Now imagine a world where banks aren’t even content to leverage their balance sheets 10-to-1, and where regulators look the other way by broadening the definition of « capital » and narrowing the definition of « assets », so that banks can risky assets at stunningly high ratios to their own capital.
Welcome to our world.
Last week, Bloomberg’s Jonathan Weil wrote the best article you’ll find on the emerging European banking crisis (see When No. 1 Financial-Strength Ranking Spells Doom ).
Weil notes that just 86 days after passing European « stress tests, » the French-Belgian bank Dexia took a government bailout to avoid collapsing. « The stress-test exercise was a charade, just as it was a year earlier when Bank of Ireland Plc and Allied Irish Banks Plc passed their tests and collapsed soon after. Once again the rules were rigged so only a handful of unimportant banks would flunk. Everyone who was paying attention understood this. »
Weil notes that although Dexia had little more than 1% in tangible equity behind its assets, « Dexia nonetheless managed to show a capital ratio of 12.1 percent. Dexia got that ratio mainly by excluding the bulk of its assets — a process speciously referred to as risk-weighting –along with billions of euros of pent-up losses on soured holdings such as Greek government bonds. The denominator in the ratio got smaller, the numerator got bigger, and Dexia wound up looking like one of Europe’s safest banks… The takeaway here is you can’t believe anything about regulatory capital benchmarks, in Europe or elsewhere, stressed or not. It’s a lesson the world should have learned long ago, yet keeps relearning. »
It would be funny, if not so distressing, that the day before Dexia failed, the Italian bank Intesa Sanpaolo presented a chart showing that it ranked among the top 4 in the European stress tests, versus 20 of its peers. The top bank on the list? Dexia.
As a side note, Bill Hester observes that despite the fact that Dexia was a fraction of the size of other European banks, its rescue last week triggered an immediate spike of about 80 basis points in Belgium’s 5-year notes. Recall also that Ireland’s deal to save its own banks is one of the main factors that has put the country on the list of Europe’s most distressed countries. It should be clear that bailing out the banking system is not as simple or clean as it might seem when the countries doing the bailouts are already experiencing fiscal strains.
Weil continues « In a way, by blowing its job so spectacularly, the European Banking Authority may have done the public a favor. Now that we have a clear point of reference, all you need to do to see what other European banks we should be worried about is look up which ones were sporting capital ratios similar to Dexia’s. For instance, as of Dec. 31, four other European banks that passed this year’s stress tests had Tier 1 capital ratios of more than 10 percent while showing tangible common equity ratios of less than 2 percent, according to data compiled by Bloomberg. France’s Credit Agricole was one. The others were Germany’s Commerzbank, Landesbank Berlin and Deutche Bank. »
As we entered 2008, I put together a spreadsheet to track financial institutions that were of particular concern based on their gross leverage (the ratio of total assets to the institution’s own capital), and the ratio of tangible equity to total assets. The most leveraged institutions at the time were Fannie Mae, Freddie Mac, Bear Stearns, Merrill Lynch, and Lehman Brothers. That spreadsheet turned out to be a fairly good predictor of the institutions that would either fail, go into receivership, or require bailouts as a result of insolvency. The corresponding calculations for several major European Banks are below. These calculations essentially mirror Weil’s list. Landesbank Berlin, Deutsche Bank and Credit Agricole are of greatest concern. While Danske Bank technically has a higher leverage ratio than Commerzbank, it has a larger buffer in the form of common equity – Commerzbank has only 1% of tangible common equity against its assets, the other 2% being more bond-like preferred equity.
When you consider the fact that most U.S. banks, just before the U.S. credit crisis in 2008, sported gross leverage ratios of about 12 (where Citigroup, Morgan Stanley, Goldman Sachs and JP Morgan remain today), the gross leverage ratios of European banks today are truly astounding.
Weil ends his piece with a simple sentence: « Dexia’s demise is only the start. » We couldn’t agree more.
Congratulations, Dr. Sargent
One of the better pieces of news last week came from Stockholm, where the Nobel prize for economics was awarded to Thomas Sargent, for his work in rational expectations. Rational expectations methods are used to analyze situations where people in the economy actually think a little bit, and cannot simply be fooled repeatedly by policymakers to believe things that constantly contradict the evidence (which is what monetary theory invariably assumed prior to the 1970’s, and which Ben Bernanke still insists upon today).
A simple, neat example of a rational expectations model is the exchange rate overshooting model, which recognizes that a monetary policy that reduces short-term interest rates but raises long-term inflationary expectations will tend to produce a large and immediate depreciation in the currency, followed by a slow appreciation, which is what we observed in response to QE2. It is important to understand though, that the methods that Sargent contributed are useful far beyond models where all individuals behave rationally. They also allow assumptions of partial or bounded rationality, but within a framework that actually allows carefully derived and important conclusions. Rational expectations methods are also critical when thinking about the long-run effects of economic policies that have the capacity to alter economic expectations and incentives.
Seemingly a lifetime ago, Tom Sargent was my dissertation advisor at Stanford, where I was fortunate to have Ronald McKinnon, Joseph Stiglitz, John Taylor and Robert Hall as my other advisors and committee members, which seemed incredible enough at the time but now seems almost bizarre in terms of the sheer firepower of that group. As important as anything I learned from them individually is what I learned from them collectively, which was the value of research and rigorous thought (which is part of the reason for my disdain for Wall Street’s toy valuation approaches like forward operating earnings multiples, which are a cheap substitute for careful discounted cash flow valuation). To underscore the fact that rational expectations methods need not assume that everyone behaves rationally, the title of my doctoral dissertation was « Market Efficiency and Inefficiency in Rational Expectations Equilibria – Dynamic Effects of Heterogeneous Information and Noise. »
In news reports of the Nobel Prize, Dr. Sargent has been characteristically humble and economical in his words, saying of himself and Christopher Sims, « We’re just bookish types that look at numbers and try to figure out what’s going on. » That’s a fairly accurate statement, but leaves out a lot of personality, and certainly understates the level of intellectual rigor that he maintains. What I remember best about our research discussions was his tendency to answer each of my questions by sending me off to work on two more. When I would try to approach an economic problem with inferior analytical tools, or partial-equilibrium thinking, he encouraged me to learn more rigorous methods before going forward, saying simply: « Remember, investment precedes consumption. »
Last week, Sargent summarized his views on the importance of thinking carefully about economic policy before acting, saying « We experiment with our models, before we wreck the world. »
Aside from Sargent’s rejection of economic policies that treat people like unthinking bricks, I frankly don’t know his political leanings, because the focus of his teaching was always on careful analysis of the whole economic equilibrium, not on any particular policy dogma. He certainly was not a particular fan of Keynesian economics, I think because he saw it as lacking any real time dimension by ignoring any consideration of the productivity or unproductivity of investment (which Keynes treats like any other class of expenditure) or any meaningful tradeoff between present and future. That said, Sargent also had a pretty dry sense of humor. I remember him showing how to write Matlab code, saying that you could insert whatever comments or documentation you like after a « % » character, after which he wrote « % Keynes was right. »
The New York Sun recently provided a recent few policy notes from Dr. Sargent. Regarding the 2009 fiscal stimulus package, he observed « The calculations that I have seen supporting the stimulus package are back-of-the-envelope ones that ignore what we have learned in the past 60 years of macroeconomic research… I recall it being said that there was widespread agreement of a big fiscal stimulus among the vast majority of informed economists. [The President’s] advisers surely knew that this was not an accurate description of the full range of professional opinion. »
On the banking system, he quoted Walter Bagehot (see Bagehot’s Rule and the Cost of Being Technically Insolvent ), who believed that central banks had the capacity to do more harm than good, and who insisted that liquidity provided during a crisis be provided at a high rate of interest and only to solvent institutions capable of providing sound collateral: « Nothing can be more surely established by a larger experience than that a Government which interferes with any trade injures that trade. The best thing undeniably that a Government can do with the Money Market is to let it take care of itself. » Sargent undoubtedly winces at the corruption of incentives that is inherent in the constant promise to make bank bondholders whole at public expense, regardless of how recklessly those banks misallocate capital.
In my view, the Nobel Prize is well deserved – there’s little question that Dr. Sargent’s work changed economics for the better. As one of his former students, there’s not a doubt that he did the same for me.
As of last week, the Market Climate in stocks remains negative, but has deteriorated significantly from the more benign negative levels that we’ve seen in recent weeks. Generally speaking, the worst market plunges tend to feature three things – overvaluation, negative market action, and a short-term overbought condition. You rarely see the three together, because establishing that sort of condition requires a strong rally against both overvaluation and negative internals. That’s about where we are, though we can’t rule out a modest extension for a bit – mostly because advisory bearishness is reasonably elevated as of last week. That said, the drop in the CBOE volatility index late last week suggests an abandonment of bearish views, and more generally, just as early shifts toward advisory bullishness at the beginning of bull markets are often accurate and followed by further gains, early shifts toward advisory bearishness at the beginning of bear markets are also often accurate and followed by further losses. Overall, market conditions remain negative, and Strategic Growth and Strategic International Equity remain well hedged. In Strategic Total Return, we continue to have about 18% of assets in precious metals shares, representing the largest driver of day-to-day fluctuations, with less than 4% of assets in utility shares and foreign currencies, and a bond duration of about 1.5 years in Treasury securities.