The first “modern model” of secular stagnation, “A Model of Secular Stagnation” by Gauti Eggertsson and Neil Mehrotra (E&M), has just been released. From the abstract:
In this paper we propose a simple overlapping generations New Keynesian model in which a permanent (or very persistent) slump is possible without any self-correcting force to full employment. The trigger for the slump is a deleveraging shock which can create an oversupply of savings. Other forces that work in the same direction and can both create or exacerbate the problem are a drop in population growth and an increase in income inequality. High savings, in turn, may require a permanently negative real interest rate. In contrast to earlier work on deleveraging, our model does not feature a strong self-correcting force back to full employment in the long-run, absent policy actions.
Successful policy actions include, among others, a permanent increase in inflation and a permanent increase in government spending. We also establish conditions under which an income redistribution can increase demand. Policies such as committing to keep nominal interest rates low or temporary government spending, however, are less powerful than in models with temporary slumps.
It appears that the Fed is being innocuous with its ‘forward guidance’ on interest rates. The preffered policies are a permanent increase in the inflation target and also a permanent increase in government spending. What are the chances of these preffered policies being enacted? Very close to nil!
What if the trigger for the slump was monetary? In that case, deleveraging is just a consequence of the monetary mistake that caused the slump. Likely, a monetary ‘revival’ would contribute to an exit from the slump.
E&M do not acknowledge a monetary problem. Like anyone trained over the last 30 years they have been taught to “ignore money”. It´s all about “Interest & Prices”! Nevertheless, if the crisis has taught us anything macroeconomic wise, it is that insisting on targeting inflation (once inflation has been conquered) is not the best idea, and that using interest rates to gauge the stance of monetary policy is even worse.
The chart illustrates the nature of the slump. It shows the depressive state of the economy (relative to where it should be). And the slump took place after a little over two decades of nominal stability! Maybe it was due to Greenspan´s ‘lucky stars’. The fact that nominal stability was lost so suddenly and abruptly gives some credence to that conjecture.
Be that as it may, some will convincingly argue that the previous trend level is not attainable any longer, but the fact remains that there surely is an attainable level higher than the one we are at.
The panel below illustrates the monetary nature of the crisis. When money demand increased (velocity fell), likely reflecting the higher uncertainty brought on by the house price bust, the Fed, wrongly thinking interest rates are a good measure of the monetary policy stance, unwittingly allowed the growth in money supply (here measured by the Divisia M3 monetary aggregate) to contract. No wonder nominal spending (NGDP) tanks.
When QE1 was introduced in March 2009, it quickly reversed the downtrend in velocity. Although money growth remained contractionary, the fall in money demand (rise in velocity) was enough to reverse the NGDP growth trend. Later money supply growth increased and over the last 4 years what we observe is that the Fed has “calibrated” (just as unwittingly as in the Greenspan era) NGDP growth at about 4%. The depressed state (responsible for the high unemployment/low employment) is due to the fact that the Fed never tried to ‘regain’ the original trend level, or a level closer to it.
So, please let´s not enshrine “secular stagnation”; much better to think about the appropriate monetary framework to regain nominal stability. Peter Ireland and Michael Belongia have just such a suggestion in A “Working” Solution to the Question of Nominal GDP Targeting. From the abstract:
Although a number of economists have tried to revive the idea of nominal GDP targeting since the financial crisis of 2008, very little has been said about how this objective might be achieved in practice. This paper adopts and extends a strategy first outlined by Holbrook Working (1923) and later employed by Hallman, et al. (1991) in the P-Star model. It presents a series of theoretical and empirical results to argue that Divisia monetary aggregates can be controlled by the Federal Reserve and that the trend velocities of these aggregates exhibit the stability required to make long-run targeting feasible.
 The Divisia monetary aggregates differ from the simple sum (M1, M2, etc) monetary aggregates in that they are weighted sums, where the weights take account of the monetary services provided by the components. For example, a checking account is more liquid than a time deposit account. Often, the information the Divisia Monetary Aggregates provide is very different from the usual aggregates.