Tuesday, July 10, 2012

Hussman: How QE Works

July 9, 2012 What if the Fed Throws a QE3 and Nobody Comes?

John P. Hussman, Ph.D.
All rights reserved and actively enforced.

Reprint Policy
The financial markets were largely unresponsive to news of further easing by the European Central Bank, the Bank of England, and the People’s Bank of China last week. Notably, Spanish bonds plunged, while German short-term government bonds now yield -0.17%, indicating growing concern about sovereign default risk in the Euro area. Every few days will undoubtedly bring word of new “agreements” and “mechanisms” – arcane enough to mask their futility – that promise to solve the European crisis. The headwinds remain very strong. The key distinction here remains liquidity versus solvency. There is little doubt that liquidity will be provided at every opportunity, though the continual degrading of collateral standards by the ECB suggests that all the good collateral has been pledged already. More importantly, with a global recession visibly unfolding, solvency risk will only increase.
The odds remain against European countries agreeing to the surrender their national sovereignty to the extent needed to create a “fiscal union” and enable massive and endless transfers of public resources from stronger to weaker European countries. Barring a catastrophe severe enough to either prompt European countries to hand fiscal control to a central administrator, or to prompt Germany to agree to unconditional bailouts, the least disruptive move would be for Germany and a handful of stronger countries to leave the Euro first, and allow the remaining members to inflate as they wish.
With regard to the economy, I noted two weeks ago that the leading evidence pointed to a further weakening in employment, with an abrupt dropoff in industrial production and new orders. Mike Shedlock reviews the litany of awful figures we’ve seen since then, focusing on the new orders component of global purchasing managers indices: U.S. manufacturing new orders and export orders plunging from expansion to contraction, Eurozone new export orders plunging (only orders from Greece fell at a faster rate than those of Germany), and an accelerating decline in new orders in both China and Japan.
Recall that the NBER often looks for “a well-defined peak or trough in real sales or industrial production” to help determine the specific peak or trough date of an expansion or recession. From that standpoint, the sharp and abrupt decline we’re seeing in new orders is a short-leading precursor of output. As the chart below of global output suggests, I continue to believe that we have reached the point that delineates an expansion from a new recession.

On the employment front, Friday’s disappointing report of 80,000 jobs created in June may be looked on longingly within a few months, as we continue to expect the employment figures to turn negative shortly. That said, it remains important to focus on the joint action of numerous data points, rather than choosing a single figure as an acid test. I noted last week in Enter, the Blindside Recession, GDP and employment figures are subject to substantial revision. Lakshman Achuthan at ECRI has observed the first real-time negative GDP print is often seen two quarters after a recession starts. Earlier data is often subsequently revised negative. As for the June employment figures, the internals provided by the household survey were more dismal than the headline number. The net source of job growth was the 16-19 year-old cohort (even after seasonal adjustment that corrects for normal summer hiring). Employment among workers over 20 years of age actually fell, with a 136,000 plunge in the 25-54 year-old cohort offset by gains in the number of workers over the age of 55. Among those counted as employed, 277,000 workers shifted to the classification “Part-time for economic reasons: slack work or business conditions.”
What if the Fed throws a QE3 and nobody comes?
To date, the stock market has largely shrugged off the evidence of oncoming recession, in the confidence that the Federal Reserve will easily prevent that outcome and defend the market from any material losses. On that point, it is helpful to remember that the real economic effects of Fed actions in recent years have been limited to short-lived bursts of pent-up demand over a quarter or two. Not surprisingly, as interest rates are already low, and risk-premiums on more aggressive assets are already remarkably thin, the impact of quantitative easing around the globe continues to show evidence of diminishing returns.
With the help of some preliminary work from Nautilus Capital, the following charts present the market gains, in percent, that followed versions of quantitative easing by the Federal Reserve, the European Central Bank, and the Bank of England on their respective stock markets (measured by the S&P 500, the Dow Jones EuroStoxx Index, and the FTSE Composite, respectively). In order to give QE the greatest benefit of the doubt and account for any “announcement effects,” the advances in each chart are based on the 3-month, 6-month, 1-year and 2-year gains in each index following the initiation of the intervention, plus any amount of gain enjoyed by the market from its lowest point in the 2 months preceding the actual intervention. The effects of most interventions would look weaker without that boost.
Remember that quantitative easing “works” through central bank hoarding of long-duration government bonds, paid for by flooding the financial markets with currency and reserves that essentially bear no interest. As a result, investors in aggregate have more zero-interest cash, and feel forced to reach for yield and speculative gains in more aggressive assets. Of course, in equilibrium, somebody has to hold the cash until it is actually retired (in aggregate, “sideline” cash can’t and doesn’t “go” anywhere). Increasing the quantity simply forces yield discomfort on more and more individuals. The process of bidding up speculative assets ends when holders of zero-interest cash are indifferent between continuing to hold that cash versus holding some other security. In short, the objective of QE is to force risky assets to be priced so richly that they closely compete with zero-interest cash.
Understanding this dynamic, it follows that QE will have its greatest impact on financial markets when interest rates and risk-premiums have spiked higher. If interest rates are low already, and risky assets are already priced to achieve weak long-term returns (we estimate that the S&P 500 is likely to achieve total returns of less than 4.8% over the coming decade), there is not nearly as much room for QE to produce a speculative run. Leave aside the question of why this is considered an appropriate policy objective in the first place, given the extraordinarily weak sensitivity of GDP growth to market fluctuations. The key point is this – QE is effective in supporting stock prices and driving risk-premiums down, but only once they are already elevated. As a result, when we look around the globe, we find that the impact of QE is rarely much greater than the market decline that preceded it.
To illustrate, each of the Fed, ECB and BOE quantitative easing interventions since 2008 are presented below as a timeline. The shaded area shows the amount of market gain that would be required to recover the peak-to-trough drawdown experienced by the corresponding stock index (S&P for Fed interventions, EuroStoxx for ECB interventions, FTSE for BOE interventions) in the 6-month period preceding the quantitative easing operation. The lines plot the 3-month, 6-month, 1-year and 2-year market gain following each intervention, adding any gain from the low of the preceding 2 months, to account for any "announcement effects." Technically, the lines should not be connected, since they represent the gains following distinct actions of different central banks, but connecting the points shows the clear trend toward less and less effective interventions, with the most recent interventions being flops. Notice also that central banks have typically initiated QE interventions only when the market had somewhere in the area of 18% or more of ground to make up.

Of all the experiments with QE, the round of QE2 from late-2010 to mid-2011 was most effective, in that stocks recovered their prior 6-month peak, and even some additional ground. Yet even with QE2, the Twist and its recent extension, as well as liquidity operations such as dollar swaps and so forth, the S&P 500 is again below its April 2011 peak, and was within 5% of its April 2010 peak just a month ago (April 2010 is a particularly important reference for us, since that is that last point that the ensemble methods we presently use would have had a significantly constructive market exposure). The largely sideways churn since April 2010 reflects repeated interventions to pull a fundamentally fragile economy from the brink of recession, and recessionary pressures are stronger today than they were in either 2010 or 2011. Investors seem to be putting an enormous amount of faith in a policy that does little but help stocks recover the losses of the prior 6 month period, with scant evidence of any durable effects on the real economy.
In short, the effect of quantitative easing has diminished substantially since 2009, when risk-premiums were elevated and amenable to being pressed significantly lower. At present, risk-premiums are thin, and the S&P 500 has retreated very little from its April 2012 peak. My impression is that QE3 would (will) be unable to pluck the U.S. out of an unfolding global recession, and that even the ability to provoke a speculative advance in risky assets will be dependent on those assets first declining substantially in value.