House on Ice
by John P. Hussman, Ph.D., Hussman Funds
"If you have bad banks then you very urgently want to clean up your banks because bad banks go only one way: they get worse. In the end every bank is a fiscal problem. When you have bad banks, it is in a political environment where it is totally understood that the government is going to bail them out in the end. And that's why they are so bad, and that's why they get worse. So cleaning up the banks is an essential counterpart of any attempt to have a well functioning economy. It is a counterpart of any attempt to have a dull, uninteresting macroeconomy. And there is no excuse to do it slowly because it is very expensive to postpone the cleanup. There is no technical issue in doing the cleanup. It's mostly to decide to start to grow up and stop the mess."
MIT Economist Rudiger Dornbusch, November 1998
On the surface, the U.S. economy is gradually recovering. Based on mean reversion to potential GDP (which generally occurs over a 4-year horizon absent an intervening recession), we would expect GDP growth over the coming 4-year period to average 3.8%, with average monthly employment growth of 200,000 jobs. This would be my "benchmark" expectation if the U.S. and international banking systems were "clean." However, my concern is that the surface U.S. recovery is built over a foundation that is vulnerable to further strains. If our policy makers had made proper decisions over the past two years to clean up banks, restructure debt, and allow irresponsible lenders to take losses on bad loans, there is no doubt in my mind that we would be quickly on the course to a sustained recovery, regardless of the extent of the downturn we have experienced. Unfortunately, we have built our house on a ledge of ice.
Debt burdens have not been meaningfully reduced in the mortgage sector - they have only been extended. Home values are still well above their historical norm relative to incomes. Yet more than 20% of homeowners already have "negative equity" - mortgages that exceed the current prices of their homes. A few months ago, Deutsche Bank projected that the negative equity rate may rise to 48% in 2011. Yet even if we ignore the mortgages that have been "modified" by slapping delinquent payments onto the back of the obligation, one in seven U.S. homes is presently delinquent or in foreclosure. As much as we have done to make lenders whole, nothing apart from a major restructuring of mortgage obligations will ease the continuing vulnerability on the debtor side.
Meanwhile, the dependence of the banking system on short-term deposits is worse than it was prior to the crisis. The FDIC reports that time-deposits have declined for the 7th consecutive quarter, while the cost of funding assets has dropped below 1%, as banks rely on the shortest liabilities possible in order to earn higher interest spreads. So while the month-to-month progress of the economy has been somewhat encouraging, our policy makers have put us in the position of continually revisiting a can that they simply kicked down the road.
As we look ahead to the coming years, I believe that the best way to avoid major losses will be to remain mindful of the distinction between surface economic progress and latent (underlying) risks. As a starting point, we'll look at the "benchmark" scenario - the potential growth in GDP and employment that we can expect in the absence of further economic shocks. Second I think it's useful to review the observations that the late MIT economist Rudiger Dornbusch made in 1998, many of which are directly applicable to the present environment. Finally, we'll review the current state of the economy and the financial markets.
The Mean-Reversion Benchmark
Although estimates by the Congressional Budget Office indicate potential GDP (driven by population growth and other factors) is likely to expand by only 2.3% annually over the coming four years, the "output gap" - the difference between actual GDP and current potential GDP - is large enough that simply closing that gap would require an additional 1.5% of GDP growth annually over the coming 4 years. In other words, simply closing the gap between actual and potential GDP over the coming 4-year period implies a "benchmark" expectation of roughly 3.8% annual real growth. This requires that we avoid fresh credit strains or other shocks.
Similarly, the pool of unemployed workers is sufficiently large, and potential gains from downsizing are sufficiently exhausted, that we would expect to observe a gradual expansion in employment simply on the basis of long-term mean reversion, provided that we avoid fresh credit strains or other shocks. The U.S. economy has lost over 7.5 million jobs in less than three years, putting non-farm payrolls at roughly 130 million jobs. Historically, the growth rate of payrolls has run at about half the growth rate of real GDP (with the remainder accounted for by productivity growth - this is a corollary to Okun's Law). So 3.8% real GDP growth implies job growth of about 1.9% annually. In an economy gradually reverting back to potential over the coming four year period, that would imply net job growth of about 2.5 million jobs annually, or about 200,000 jobs a month, on average.
As a side note, even if you tack on 2.2% inflation to get 6% growth in nominal GDP and corporate revenues, stock market investors are still stuck with the fact that any normalization profit margins and profits to GDP (which are highly cyclical and presently near their historic peak) would result in roughly zero earnings growth for the S&P 500 over the 4-year period.
These "benchmark" estimates assume that we avoid another economic downturn in the next few years. In the chart below, you can see the impact of such downturn on the actual 4-year growth of the economy, versus what one would have projected. Notice that in the chart below, periodic recessions have caused GDP growth to fall short of the benchmark expectations. For instance, the 1981-1982 recession caused a shortfall versus the 4-year growth that one would have anticipated in 1977-1978. The 1991 recession caused a shortfall versus what one would have expected in 1987. And the 2009 collapse caused a shortfall versus what one would have expected in 2005.
So the current "benchmark" expectation for 4-year real GDP growth is about 3.8% based on a gradual reversion to potential GDP, and the corresponding "benchmark" for job growth is about 200,000 jobs per month on a sustained basis. Still, these levels are highly dependent on avoiding a fresh recession in the interim, and unfortunately, the U.S. is perpetuating policy mistakes that threaten that result.
Hard Money and Clean Banks
In 1998, MIT economist Rudi Dornbusch gave a set of lectures in Munich on "International Financial Crises." I've excerpted some of his remarks below. Much of what Dornbusch discussed is particularly relevant to the present credit strains in Europe and various small countries across the globe, but is also important with respect to how we continue to approach difficulties in the United States. Dornbusch highlighted three essential concerns: 1) the vulnerability of banking systems that are dependent primarily on short-term deposits and funding that can be withdrawn on demand, 2) the risk of having national liabilities denominated in a foreign currency (which is essentially the case with Ireland, Portugal, Greece and Spain, all of which are constrained by the European Monetary Union and cannot simply print their own money), and 3) the importance of restructuring bad debt and avoiding bank bailouts.
"So let me focus on these new financial crises and ask where do they come from? It's very easy to predict a crisis, but then you have to wait for two years until it happens. There is the issue of contamination. Why innocent bystanders get hit. And there is the last issue of the depth of crisis. Why are they so traumatic, leaving a crater that is unbelievably deep?
"The argument is that the national balance sheet is extremely vulnerable. You could live for years and years and years with a balance sheet like that, and nothing ever happens, and nobody ever talks about you, and you are a great story of success, with high growth and a miracle. In a very rich country you can afford to do bad things very, very long.
"And then something happens and then it turns out that that very vulnerability is such that it's a dramatic end of all success. Suddenly in the afternoon of an otherwise sunny day, the world financial system wants 50 billion dollars from a country, and the country doesn't have the 50 billion dollars, and the IMF cannot get there fast enough, and the next thing, the place blows up, and there is a massive depreciation of the currency, pervasive bankruptcy, and the fire spreads to the next country.
"A banking system is a very important part of how a country hangs together... Very, very large, very, very short term liabilities can in no time become a bankruptcy issue for an entire banking system, more so if it's unregulated. And that means people can want to get their money back tomorrow afternoon, and when someone wants their money back, they all want their money back. And if they all want their money back, there is no way for an economy to pay at short notice. And if I can't repay, then they'll be much more eager to get their money, and as the bank run occurs, of course the rest of the economy will collapse with it. Maturity is the first issue.
"The second issue is denomination. There is a very big risk for a country to denominate its liabilities in foreign exchange. Something that they cannot print, something that they cannot get their hands on, and therefore something that is very vulnerable if in fact the exchange moves, the burden of those liabilities increases and bankruptcy of the country and the underlying companies becomes a big issue. Because if that is seen to be happening, then of course, everybody wants their money before it happens, and as they try to get it, they provoke the very collapse that I'm describing.
"If a country has an extraordinary withdrawal of short-term credit and they made the mistake of having short-term credit positions, then they should have a debt restructuring. Then they should say, 'We are now going to default on our debt and we are very, very sorry and what you thought was an overnight loan actually isn't an overnight loan - it is a long-term stabilisation loan.' That makes an extraordinary important distinction between direct investment, which should be favoured at all times, and you should create a history of always treating direct investment extremely well, and short-term debt, which if things go bad may turn out to be long-term debt.
"If you have bad banks then you very urgently want to clean up your banks because bad banks go only one way: they get worse. In the end every bank is a fiscal problem. When you have bad banks, it is in a political environment where it is totally understood that the government is going to bail them out in the end. And that's why they are so bad, and that's why they get worse. So cleaning up the banks is an essential counterpart of any attempt to have a well functioning economy. It is a counterpart of any attempt to have a dull, uninteresting macroeconomy. And there is no excuse to do it slowly because it is very expensive to postpone the cleanup. There is no technical issue in doing the cleanup. It's mostly to decide to start to grow up and stop the mess.
"If you have a hard money and if you have clean banks then you don't have macroeconomics as a problem anymore. Yes, you have slowdowns in the economy and yes you have booms and the key macro problem, the government, has been taken out of it. That's very important to understand that in the economies we are talking about, the problem is the government. The government is not the solution. Hyperinflations are made by governments, debt defaults are made by governments, exchange rates crises are made by governments. And if they don't know how to do it well, and the assumption is: no they do not know how to do it well, then take them out of the business."
Dornbusch likens an unsound financial system to drunk driving. "Think of someone who has made a great expertise of drunk driving, regularly drives drunk, tells you that he never has a problem, and one day there is a terrible, terrible accident. And he'll say, âWell, it was the red light. It wasn't my being drunk. Normally that light is green.â Drunk driving, which for years has worked, with a financial structure that is recklessly, recklessly unsound. But the light was green and then one day it wasn't green, and then the house of cards came crumbling down."
In applying Dornbusch's observations to the recent financial crisis, it is important to distinguish liquidity from solvency. During a bank run, it is essential to provide sufficient liquidity to ensure that depositors can get their money back. That is the only way to stop the run. But if the government does that, the only thing it should buy outright when providing liquidity is its own Treasury debt - everything else should be on a repurchase basis. The Treasury can even provide capital provided that its claims are senior to those of the bank's bondholders. U.S. policy makers did some of this correctly during the recent crisis, but they also bought bad mortgage debt outright, suspended disclosure, and avoided every attempt to restructure, which stemmed the bleeding without addressing the underlying problem.
As I've frequently noted, even if a bank "fails," it doesn't mean that depositors lose money. It means that the stockholders and bondholders do. So if it turns out, after all is said and done, that the bank is insolvent, the government should get its money back and the remaining entity should be taken into receivership, cut away from the stockholder liabilities, restructured as to bondholder liabilities, recapitalized, and reissued. We did this with GM, and we can do it with banks. I suspect that these issues will again become relevant within the next few years.
The present situation
Europe will clearly be in the spotlight early this week, as a run on Irish banks coupled with large fiscal deficits has created a solvency crisis for the Irish government itself and has been (temporarily) concluded with a bailout agreement. Ireland's difficulties are the result of a post-Lehman guarantee that the Irish government gave to its banking system in 2008. The resulting strains will now result in a bailout, in return for Ireland's agreement to slash welfare payments and other forms of spending to recipients that are evidently less valuable to society than bankers.
Essentially, the problem in Ireland is exactly what Dornbusch described: First, Ireland has a banking system that like other countries around the world, including the United States, carries a mountain of bad long-term debt on the asset side, and has become increasingly dependent on funding them with short-term deposits over the past decade, thanks to the allure of "cheap" money at the short-end of the maturity curve.
Next, Ireland and other countries in the European Monetary Union have their liabilities denominated in a currency (the euro) that they cannot actually print on their own. As Ireland, Greece, Portugal and other European countries run budget deficits, they have to induce the private market to buy their government bonds, which are denominated in the common currency. This is effectively like being on a fixed exchange rate or a gold standard, so rather than being able to print money or depreciate the currency, the only adjustment variable is the interest rate. So rates have been soaring in these countries. To some extent, states and municipalities in the U.S. are in a similar situation.
Over the short run, Ireland will promise "austerity" measures like Greece did - large cuts in government spending aimed at reducing the deficit. Unfortunately, imposing austerity on a weak economy typically results in further economic weakness and a shortfall on the revenue side, meaning that Ireland will most probably face additional problems shortly anyway.
Germany's Chancellor Angela Merkel is effectively the only major leader who recognizes the correct prescription, which is - as Dornbusch advises - to grow up, restructure the debt, and clean up the banks, because bailing them out will simply make the problem worse down the road. Merkel calls this "burden sharing" - which is another phrase for "restructuring" - but she is also vilified for it, because lenders would much prefer to have the government make them whole at public expense (and mostly the German public at that). And so, predictably, Europe is now choosing to kick the can down the road.
Here at home, the situation is only a little different from the standpoint of underlying fundamentals, but as noted at the outset, month-to-month economic progress has been reasonably positive in the U.S. lately, so there is a larger distinction between surface conditions and latent ones.
The main problem, of course, remains mortgage debt. Unfortunately, because of the suspension of FASB mark-to-market rules, investors don't actually have the ability to fairly assess the balance sheet of the banking system. Still, various mortgage statistics provide a partial, if incomplete picture. The latest report from Lender Processing Services notes that the inventory of homes in foreclosure is now 7.4 times the historical average, and rising. When a home actually enters foreclosure, it is removed from "delinquent" status, so distressed housing can shift from "delinquent," to "foreclosure," to "real-estate-owned" classifications depending on where they are in the chain, which is important when interpreting these statistics.
Though about 20% of homes that have been delinquent for more than two years are still not in foreclosure, distressed mortgages eventually either go into foreclosure, or they are "cured." What is interesting about this is that a growing portion of homes classified as newly delinquent are actually re-defaults of homes that were previously delinquent and were temporarily "cured" through modification. Typically, these modifications involve taking all of the missed payments (which in more than one-third of the cases have been delinquent for well over a year), and tacking them on to the back-end of the loan, essentially extending the maturity, and often reducing the interest rate for the first year of the modified loan. Because of this repeated cycle of cure and re-default, the number of distressed properties is most likely significantly higher than the still-elevated delinquency and foreclosure rates might suggest.
From my perspective, this is another perpetual game of kick-the-can-down-the-road to prevent mortgages from being classified as delinquent, which prevents banks from having to reserve against loan losses or write down the value of the assets. The real question is whether this is sustainable. Since perpetuating this Ponzi scheme seems to be official U.S. policy at every level, further strains may follow Dornbusch's law: "The theorem is that financial crises take much, much longer to come than you think and then they happen much faster than you would have thought. So you have a chance to be wrong twice."
That said, we've observed a gradual improvement in a variety of economic indicators in recent months, particularly in new unemployment claims. While some of the regional Fed indices (Philly, Empire State) have enjoyed positive surprises, the Chicago Fed survey, which is national, was disappointing, and at a level consistent with GDP growth of about 1.25%. That is still positive, but the confidence interval easily includes zero, so we're not quite to the point where we would conclude that a fresh economic downturn is "off the table." Financial strains tend to come on abruptly. Until we observe debt restructuring and transparent accounting rules (especially some modified version of mark-to-market), it will be dangerous to think of economic risks as being "off the table," when they are probably just hidden under a napkin.
Research Update
In the meantime, we have to work with the economy and the markets that we have. As I've noted in recent weeks, our present defensiveness is not driven by concerns about fresh credit or economic difficulties, but rather by a combination of overvalued, overbought, overbullish, rising-yield conditions that has generally turned out badly in post-war data. While we will remain mindful of underlying economic risks, I do believe that there is a wide enough range of outcomes in post-war data to allow us to manage the residual economic risks we face. As we clear various unfavorable elements of the current market environment, we will be able to periodically accept small or moderate exposures to market fluctuations, even in the absence of undervalued conditions. For now, however, we remain defensive on the basis of conditions that have rarely worked out well for stocks.
Our investment approach is fairly straightforward - accept proportionately greater exposure to risk when the expected return per unit of risk is high, and proportionately reduce exposure to risk when the expected return per unit of risk is low. The details are in the implementation, and that is where we focus most of our research. Over the past two years, most of these efforts have focused on the proper use of multiple data sets, on broadening the range of classifications that we define as distinct "Market Climates," and on enhancing the "robustness" of these classifications across subsets of the data.
Without going into technical details that would be useful only to competitors, the basic outcome of this research is that we continue to refine the Market Climates we identify, and have developed additional methods to make robust estimates of their associated return/risk profiles. I expect that you'll observe the first fruit of this research in the form of modest, transitory exposures to market fluctuations on a more frequent basis. This isn't a major change - large, persistent exposures will still await some combination of significantly improved valuations, downward yield pressures, and economic clarity - but we should be able to better exploit our latitude for modest variations in our market exposure as conditions vary over time. As always, we retain a strong emphasis on risk management, with the objective of outperforming our benchmarks with smaller periodic drawdowns than a passive investment strategy.
I continue to view the stock market as richly valued, and priced to achieve returns of less than 5% at every horizon out to a decade. The expected returns at the shorter horizons are more volatile, of course, than those at longer horizons, and it is there that a broader range of "Market Climate" classifications can be helpful. We'll modestly alter our exposure to market fluctuations in response to modest changes in conditions, but of course, we would prefer a large shift in valuations which would allow us to accept an unhedged exposure.
Here and now, we remain tightly defensive. We can certainly identify conditions under which one could have expected a speculative benefit, on average, from taking market risk despite overvaluation. But when overvaluation has been coupled with overbought, overbullish, rising yield conditions, the average outcomes have been quite poor.
Market Climate
As noted above, the Market Climate in stocks remains characterized by overvalued, overbought, overbullish, rising-yield conditions that have historically been associated with poor market returns. Strategic Growth and Strategic International Equity remain tightly hedged at present.
In recent weeks, we've observed a decided tendency toward "risk on" and "risk off" days, where entire classes of securities are treated as if they were a single object. During "risk on" days, the market advances, paced by financials, cyclicals, commodities, and smaller speculative issues, while the dollar weakens. During "risk off" days, the market declines, with relative strength in consumer, health, and high-quality stocks, while the dollar rallies. Since our stock selection generally focuses on higher quality issues, which has served us very well over the long-term, this type of pointed "risk on/risk off" behavior creates a situation where our stocks appear to have a smaller "beta" on a day-to-day basis than we actually believe is applicable over large moves or longer horizons.
As a result, the equity funds may respond slightly counter to general market movements on a day-to-day basis. While I expect that this is short-term behavior, we are still gradually modifying our hedge ratios in response. We are doing this carefully, since it is likely that our stock holdings will pop back to having their normal, full betas in the event of a serious market decline. Suffice it to say that our emphasis on what we view as higher quality stocks (e.g. stable revenue growth, profit margins and balance sheets) makes us a bit more sensitive to the recent risk on/risk off pattern of market action, and while we see this primarily as local, day-to-day behavior, we are gradually modifying our hedge ratios accordingly.
