Cycle drag or stagnation theory for economy?
“We meet again, at the turn of the tide. A great storm is coming, but the tide has turned.”
J.R.R. Tolkien, The Two Towers
Atide that never turns? This is what the secular stagnation hypothesis (almost) conveys in the limit: the demand deficiency is so deeply entrenched, the argument goes, that it may take a long time to remedy its causes and consequences. Hence, interest rates should – and will – remain near zero, even in the negative territory. A more moderate version would predict at least zero real interest rates for a time to come. We have a lot of concepts – and accompanying models and measurements – aiming at explaining the sluggish recovery post 2007-09. We have the debt super-cycle (Rogoff ), the financial cycle drag (BIS) and the secular stagnation (Summers) at the very least. The common factor is that all of them take for granted that markets are not self-equilibrating, that a long-view approach should be adopted, and that models and ideas should be refreshed and reinvigorated.
Secular stagnat on hypothes s
The secular stagnation hypothesis culminates in the claim that demand was already structurally deficient before Lehman. Aggregate demand was lacking, because there was already insufficient investment demand in machinery and equipment, possibly because there was technological innovation that had rendered some sorts of classical investments superfluous, the last 40 years’ rising income inequality and aging population. So, aggregate demand weakness and excessive financialization would have led to a permanent interest rate fall. The upshot is: the equilibrium interest rate will remain low, and maybe even near zero for an indefinite period. Because the claim that demand will remain low because it is built-in the system of the last decades is the cornerstone of this thesis, the consequences of the 2007-09 crisis may not be remedied easily, and would require low interest rates until demand finally redresses itself. One could even trace the origin of the malaise back to the 1973 Oil Shock, after which the New Deal was abandoned, and inequality rose.
Close gap support mechan sm des gns only part ally worked
However, there is also the possibility Rogoff raised two years ago that the credit smile is biased against high-risk borrowers who could have pushed loan interest rates up. So, interest rates remained low for a few years not because demand cannot be triggered, but because the credit surface is different now. The fact that there was almost complete credit rationing in the aftermath of Lehman is undeniable, and this fact underlies many actions taken up by the European Central Bank (ECB) since Mario Draghi became president. Yet many a European bank refrained from extending the credit surface to SMEs and other high-risk types, even after the initial pushup that benefited both banks and governments through easy funding so sovereign debt rates could fall. The targeted credit support mechanism designs have only partially worked, and rather slowly. That might indeed explain part of the sluggishness in the rates of interest. At one time in October 2014, the extant credit stock of the Euro system was even below its nominal value of three years earlier, let alone any credit extension. The extreme de-leveraging that occurred through European banks’ strictly risk-averse behavior at around the time of Draghi’s appointment – even afterwards – cannot be easily forgotten.
A decept vely s mple theory compl cated by the real world
There is another argument, which purports to explain the persistence of low interest rates, even negative, in the aftermath of a deep financial crisis that also drags down the real economy. The argument is deceptively simple in its principle, but gets complicated when it comes to real-world computations. It dates back to a much earlier paper of Kenneth Arrow, but revived by many in the context of climate change, especially after the Stern Review of 2007. What is the correct discount rate in evaluating a long-term project?
A paper by Martin Weitzman elevated this issue in a more general context, and related it to asset prices, stock-bond disconnect, and so on. It reads roughly as follows: in the long-run risk perception –captured by the discount rate – is somewhere between the risk-free rate and the overall market rate. Now, if the market rate is represented by average real equity returns over a sufficiently long span of time, and if the short-run bond rate is a sufficient statistic for the risk-free rate, the US data would yield 7% for the general and 1% only for the risk-free rate.
The difference is huge: it would yield a present discounted value 8,000 times higher should 1% be used as opposed to 7% per annum over a span of 150 years. Now, that kind of logic should best be applied to either monumental events such as climate change or to long-horizon large public projects. However, this deceptively simple idea can also deliver an insight: after tail-risk events such as Lehman, if the market doesn’t believe in the recurrence of the said event or suchlike, the risk perception might decline so steeply that the risk-free rate could all of a sudden seem normal for the whole market risk.
The lower bound might apply as a panacea for all risks, and stock prices climb along with other asset prices whereas bond rates remain very low for an extended period of time. This is not the same as the secular stagnation hypothesis, which related such persistence to a chronic deficiency of aggregate demand and to supply-side factors – such as demographics – or to the rising inequality in the wealth/income distribution, but it relates the persistence of low rates directly to the perception of risk. Whether the weak but always latent presence of yet another tail-risk is masked by successive rounds of quantitative easing is another question.
F nanc al cycle drag hypothes s
The financial cycle drag hypothesis – or the debt super-cycle hypothesis for that matter – differs from the secular stagnation hypothesis in at least one crucial predictive aspect. The equilibrium real rate of interest must be positive because demand is not chronically deficient, and in fact it has begun to recover. Nevertheless, in my opinion what matters most is the claim that the financial and real cycles are not coincident: the financial cycle can last longer, is more often than not deeper, and it has a persistent negative effect on productivity.
Accordingly, the financial cycle drag hypothesis conjectures that the credit boom misallocates resources, and overfeeds the construction and real estate development businesses at the expense of sectors that are conducive to technological innovation and productivity catch-up. It was widely claimed in the early decades of the 20th century that the financial and real sectors had merged into one, the so-called ‘finance capital,’ and had led the way to over-concentration, cartel formation, monopolistic tendencies, capital exports to third countries and so on. The resulting business cycle would be one and the same cycle: its phases would coincide almost one by one.
The financial cycle drag says otherwise: finance acts in relative autonomy, and carries the real sector with it. The recovery after a financial cycle is slower than the usual (real) business cycle because an abnormal credit expansion and overhang liquidity predates the turn of the tide, i.e. in the boom phase misallocation takes root. As a consequence, since credit is already tied, during the recovery period it is the relatively unproductive sectors that continue to drive the economy.
What of the commod ty super-cycle dr ven by Ch na?
If aggregate demand was already weakening before 2007-09 as the secular stagnation hypothesis entails, then why was there a commodity super-cycle driven by China and other Asian countries in the first place? There was a lack of supply at that time, for instance for copper. Also, energy futures became widespread, and after 2005 that has also led to the change in the ‘volatility smile’ of oil contracts. That there was already deficient demand at the global level is also not congruent with the post-Lehman phase when the so-called ‘two-speed recovery’ had spearheaded emerging economies as bulwarks of growth. Instead of global aggregate demand deficiency, it is perhaps better to allude to the possibility of an asymmetric cycle relative to some advanced countries. They may have experienced some kind of demand weakness, an idea reminiscent of new Keynesians or Keynes himself, and before him Rosa Luxemburg.
Is the whole post-Lehman per od an over-extended cycle?
On the other hand, even the seemingly self-evident concept of cycles come in many a variety. Confining ourselves to only the outcomes of a cycle, what kind of cycle are we referring to? Is the recovery conducive to a longterm historical mean – maybe 20 years or 50? Is mean-reversion in the interest rates, if they revert to a mean anyway, conducive to the peak level of the past cycle? Is the cycle symmetric, so we will in the end see that US 10-year bond rates will climb to 4-5%? If so, either other variants will also tend to the pre-Lehman parameterization, or risk perception will change and do the job. Or, is it more reasonable to conceptualize a series that is not recurrent, a cycle that is not symmetric, an interest rate path that will not revert back to an attractor, but a path that is different? Somewhere in between perhaps. If so, the behavior of major central banks makes more sense. They not only take into account the asset-pricing channel and try to smooth the return to normalcy, but may also see the whole post-Lehman period as an over-extended cycle, which will eventually lead to an entirely different parameter space. Maybe 3% is the new upper limit for US 10-year bonds, and if so emerging markets may still benefit from the slowly turning world that may not turn completely upside down in the end.
Global interest rates may never reach the old peak levels. Neither will they stay at near-zero real rate levels. They will most probably rise in 2018 with the Fed taking the lead. However, there are many factors that are bound to shape them: new technology/new demand/ new jobs; the change in the inflationary impact of energy inputs; the coordination of central banks and asset prices. Monetary policy may remain endogenous to some extent for some time to come.