Excerpt from the Hussman Funds' Weekly Market Comment (8/25/14):
When the majority of Americans examine the world around them, they see a stock market at record highs and modest apparent improvement in the economy, but they also have the sense that something remains terribly wrong, and they can’t quite put their finger on it. According to a recent survey by the Federal Reserve, 40% of American families report that they are “just getting by,” and 60% of families do not have sufficient savings to cover even 3 months of expenses. Even Fed Chair Janet Yellen seemed puzzled last week by the contrast between a gradually improving unemployment rate and persistently sluggish real wage growth.
We would suggest that much of this perplexity reflects the application of incorrect models of the world.
Before the 15th century, people gazed at the sky, and believed that other planets would move around the Earth, stop, move backwards for a bit, and then move forward again. Their model of the world – that the Earth was the center of the universe – was the source of this confusion.
Quantitative easing and short-term interest rates
First, the following chart shows the relationship since 1929 between the monetary base (per dollar of nominal GDP) and short-term interest rates. This is our variant of what economists call the “liquidity preference curve,” and is one of the strongest relationships between economic variables you’re likely to observe in the real world. After years of quantitative easing, the monetary base now stands at 24% of GDP. Notice that less than 16% was already enough to ensure zero interest rates, so the past trillion and a half dollars of QE have done little but increase the pool of zero-interest assets that are fodder for yield-seeking speculation. Notice also that unless the Fed begins to pay interest to banks on their idle reserves, the Fed would have to contract its balance sheet by about $1 trillion just to raise Treasury bill yields up to a fraction of one percent. So the primary policy tool of the Fed in the next couple of years will likely be changes in interest on reserves (IOR). Get used to that acronym.
Interest rates and economic growth
Following this very strong economic relationship, let’s examine a very weak one. It should give pause to those who believe that low interest rates are naturally associated with stronger economic activity. The fact is that when productive opportunities to invest are available, strong economic activity and normal or even elevated interest rates are completely compatible, and those higher interest rates also raise the bar in a way that discourages borrowing for unproductive activities.
When “natural” demand for loans is weak and legitimately productive opportunities to invest are not available, there is a temptation – clearly enticing to the Federal Reserve – to attempt to artificially stimulate demand for loans in hopes of provoking greater economic activity. The problem with repressing the financial system this way is that it lowers not only bar on loan cost, but also the bar on loan quality. This is essentially what produced the global financial crisis. Artificially depressed interest rates punish savers and cause them to seek yield by channeling funds to more and more speculative areas of the economy, while encouraging already indebted borrowers to take on more debt so long as the debt can be serviced for now.
The real Phillips Curve and the phony one
Next in the parade of overly simplified relationships is one that might be pulled directly from any randomly-selected thought bubble floating out of a window at the Federal Reserve: the Phillips Curve.
What’s interesting about the Phillips Curve is that when A.W. Phillips wrote his original Economica paper in 1958 - The Relation between Unemployment and the Rate of Change of Money Wage Rates in the United Kingdom, 1861-1957 – he was looking at two things: one being wage inflation, and the other being wage inflation in a century of data where the United Kingdom was on the gold standard and general prices were quite stable. So what Phillips actually found, and what can be replicated in data across the globe, is a relationship between unemployment and real wage inflation. It’s actually a very straightforward scarcity relationship: when labor is scarce (unemployment is low), its price tends to rise relative to the price of other goods in the economy. The reverse is true when labor is plentiful. Here’s what that relationship looks like in data since the mid-1970’s (using average hourly compensation divided by the GDP deflator). It’s not a tight curve, but it essentially reflects what Phillips observed.
The winner-take-all economy
Several factors contribute to the broad sense that something in the economy is not right despite exuberant financial markets and a lower rate of unemployment. In our view, the primary factor is two decades of Fed-encouraged misallocation of capital to speculative uses, coupled with the crash of two bubbles (and we suspect a third on the way). This repeated misallocation of investment resources has contributed to a thinning of our capital base that would not have occurred otherwise. The Fed has repeatedly followed a policy course that sacrifices long-term growth by encouraging speculative malinvestment out of impatience for short-term gain. Sustainable repair will only emerge from undistorted, less immediate, but more efficient capital allocation.
In recent years, the U.S. has experienced a collapse in labor participation and weak growth in labor compensation, coupled with an increasingly lopsided distribution of whatever benefits the recent economic recovery has generated. This is not well-explained by Phillips Curves or simplistic appeals to "insufficient demand," and it is unlikely to be improved by endless monetary “stimulus” (the targets that clearly occupy the Fed’s thinking). While our economic challenges can be largely traced to more than a decade of persistent Fed-enabled misallocation of capital, most of the costs of this misallocation have fallen on labor because of a) shifting composition of labor demand that has resulted from an increasing share of international trade with countries with heavy populations of relatively unskilled labor; and b) economic features that increasingly create a “winner-take-all” distribution of economic gains.
One of the key results in international economics is that as trade opens up between nations, those nations will tend to export goods that intensely use the factor (skilled labor, unskilled labor, capital) with which they are relatively well-endowed. In a world where we have opened up trade with countries that are densely populated with unskilled workers, and where the U.S. is relatively better endowed with skilled labor and capital, the result has been something of a hollowing out of the middle class among households that aren’t themselves endowed with skilled labor or capital. Essentially, the U.S. obtains the services of low-skilled labor more cheaply from abroad than domestically. There is no great debate on this point.
Meanwhile, transfer payments like welfare and unemployment compensation allow many households to maintain consumption despite being out of those jobs, and given the ability of households to take on debt, even if they are actually living paycheck to paycheck, the produced goods get purchased, companies make a profit, government runs a deficit, the Fed keeps interest rates low which allows all the debt to be serviced, and everyone is pleasantly, if unsustainably, happy. That’s particularly true as long as nobody asks how the debt will be repaid, which is certainly what Fed policy encourages.
Rainforests and resilience
How can policy makers help to build the economy from the middle-out, and slow the both the unproductive diversion and the lopsided distribution of resources in our economic system?
First, we should begin by stopping the harm. Quantitative easing will not help to reverse this process. The dogmatic pursuit of Phillips Curve effects, attempting to lower unemployment with easy money, has done little to materially change employment beyond what is likely to have occurred without such extraordinary intervention, but has contributed to speculative imbalances and an increasingly uneven income distribution. Indeed, Fed policy does violence to the economy by helping to narrow it to what complex systems theorists call a “monoculture.” Nearly every minute of business television is now dominated by the idea that the Federal Reserve is the only thing that matters. Meanwhile, by pursuing a policy that distinctly benefits those enterprises whose primary cost is interest itself, the Fed’s policies have preserved and enhanced too-big-to-fail banks, financial engineering, and speculative international capital flows at the expense of local lending, small and medium-size banks and enterprises, and ultimately, economic diversity.
In his book, Hidden Order, complex systems theorist John Holland (who originated the use of genetic algorithms in computing) observed that despite extremely poor soil in a tropical rain forest, even a single tree can harbor over 10,000 different species of life because “the forest departs from the simple version of a food web, in which resources are only passed upward to the top predator. Instead, cycle after cycle traps the resources so they are used and reused before they finally make their way to the river system. Agents that participate in cyclic flows cause the system to retain resources. The resources so retained can be further exploited – they offer new niches to be exploited by new kinds of agents. Parts of a complex adaptive system that exploit these possibilities, particularly parts that further enhance recycling, will thrive. Parts that fail to do so will lose their resources to those that do. This is natural selection writ large. It is a process that leads to increasing diversity through increasing recycling.” Thriving, resilient economies work the same way. Monocultures are the exact opposite.
Broken links in the chain of cause and effect
The Federal Reserve’s prevailing view of the world seems to be that a) QE lowers interest rates, b) lower interest rates stimulate jobs and economic activity, c) the only risk from QE will be at the point when unemployment is low enough to trigger inflation, and d) the Fed can safely encourage years of yield-seeking speculation – of the same sort that produced the worst economic collapse since the Depression – on the belief that this time is different. From the foregoing discussion, it should be clear that this chain of cause and effect is a very mixed bag of fact and fiction.
To be fair, we do believe that some components of the Fed’s thinking are well-supported by economic evidence. For example, in her presentation at Jackson Hole, Fed Chair Janet Yellen observed that real wage inflation remains low, and that this is an indication of ongoing slack in the labor market that isn’t well-captured by the rate of unemployment. On this point, we would completely agree. To the extent that the true Phillips Curve (which relates unemployment to real wage inflation) describes reality, it’s sensible to assert that low real wage inflation informs us that the unemployment rate has not declined to a level that reflects labor market scarcity – though we should also recognize that real wage growth would already be much higher if there was not such an extreme gap between real wage growth and productivity growth.
Where we differ from Chair Yellen is in a variety of supposed cause-effect relationships that aren’t supported by evidence to any meaningful extent, and in the neglect of systemic risks that are undeniable if one has been paying any attention at all to the macroeconomy over the past 15 years.
Let’s summarize the links in the chain of cause and effect. First, as demonstrated above, it’s clear that increasing the monetary base relative to nominal GDP will predictably and reliably lower short-term interest rates. This is true at least until the point that, as has occurred across history and across countries, inflation picks up rather uncontrollably – often following a supply shock coupled with government deficit spending – with very little at all to do with the prevailing unemployment rate. In any event, nearly a century of data provides very clear evidence of a tight link between monetary base/nominal GDP and the level of short-term interest rates.