Technique without history

In a recent presentation (wonkish) Jason Furman, head of Obama´s CEA (Council of Economic Advisers), wonders about “Whatever Happened to the Great Moderation?“:

In the wake of the Great Recession, it is worth reassessing the Great Moderation hypothesis and understanding what it means for policy going forward. Was the Great Moderation hypothesis spectacularly wrong, and did researchers miss the fact that the economy was increasingly unstable? After all, in addition to the Great Recession in the United States, we have also seen a number of serious banking and exchange rate crises in countries around the world over the last few decades. On the other hand, a number of key data series have exhibited a high degree of consistency and stability since the recovery began in mid-2009, and we are now two months away from what would be the longest streak of private-sector job growth on record. Is there a sense in which the Great Moderation has continued or returned? Even if we still see low volatility in the summary statistics we use to assess the Great Moderation, does this tell us something meaningful about the economy, or does it tell us more about the shortcomings of these summary statistics themselves?

In my remarks today, I will first explore what the original results on the Great Moderation look like with an additional ten to fifteen years of data, including the Great Recession. I will also use these data to explore whether the factors that led economists to identify a Great Moderation are still present in the economy today, and whether the additional data affect our view of these factors. Second, I will sketch out some major problems with the Great Moderation hypothesis that have been highlighted by the Great Recession. Third, I will talk about why economic stability matters. I will end with a brief outline of the unfinished agenda to promote macroeconomic stabilization, focusing on areas outside of monetary policy that play an important but sometimes underappreciated role in fostering macroeconomic stability.

At the conclusion he writes:

But the Great Recession, at the very least, put the declaration of victory on the Great Moderation in substantial perspective. There is much more to macroeconomic stabilization than smoothing quarterly or annual fluctuations—the ultimate goal is to address the largest and most persistent fluctuations. In the case of the Great Recession, policy partly failed to do that, although the fact that we avoided a second Great Depression is a testament to improvements in macroeconomic and financial policy.

In one section of the paper Furman poses the following question: “Is the Great Moderation Due to Reduced Shocks or Reduced Propagation of These Shocks?”

After a technical discussion he cites Stock & Watson´s conclusion from 2003:

“The moderation in volatility is attributable to a combination of improved policy (10-25%), identifiable good luck in the form of productivity and commodity price shocks (20-30%), and other, unknown forms of good luck that manifest themselves as smaller reduced-form forecast errors (40-60%).”

If you do the arithmetic you´ll find that some form of “good-luck” is responsible from anything between 60% and 90% of the moderation in volatility that characterizes the Great Moderation!

To his credit Bernanke argued:

As Ben Bernanke (2004) pointed out, more predictable monetary policy could lead to smaller measured “shocks” for a range of reasons, including fewer monetary disturbances and more anchored inflation expectations, as well as changes in wage and price setting institutions.

To help illustrate this point, Bernanke cited work showing that seemingly exogenous “shocks” to oil prices in the 1970s could be in part traced to earlier monetary policy decisions. He also cited a series of papers showing that stable inflation expectations reduce the impact of exchange rate fluctuations. To the extent this factor contributes to smaller and less volatile shocks in an econometric model, we certainly would not want to attribute it solely to “good luck.”

Interestingly, almost two years ago in a post commenting a post from Liberty Street (New York Fed blog) I wrote:

I believe it was not the “nature” of the shocks that changed, but how “good policy” was “good” exactly because it was managed in such a way as to contain the propagation of the shocks, not because it showed “greater vigilance on inflation”. For example, no one can say that the Fed was not paying attention to inflation in the first half of 2008 (just read the statements). Things went sour exactly because the Fed was paying too much attention to “inflation”, letting monetary policy tighten so much that aggregate demand plunged. And so did inflation! The rest is history. [It appears Bernanke forgot about his previous arguments!]

Since the “good-luck” hypothesis was introduced, my question has been: Why did “good-luck” befall the economic system at the time it did, and why did the streak of “good-luck” last for so long? Also, why the “good-luck” suddenly vanishes in 2008?

The title of my post from almost two years ago was “Please, less technique and more substance”. A bit of history certainly helps here. The first chart shows the progression of real output (RGDP). Up to 2007 it fluctuates along a stable growth trend. Note the reduction in the volatility of growth around the trend during the Great Moderation. Note also the permanent(?) drop in the LEVEL of the series in 2008.

Technique w-o history_1

The second chart shows prices and inflation (PCE-Core). Isn´t it suggestive that the Great Moderation started off exactly at the time the Fed brought inflation down and continued while it was kept low and stable? No one has ever proposed that the substantive drop in inflation was mostly due to luck!

Technique w-o history_2

The third chart is the “clincher”. The Great Moderation begins exactly at the time monetary policy is successful in providing nominal stability, i.e. keeps nominal spending (NGDP) growing in a stable manner along a level path and ends at the exact time the Fed allows nominal spending to crash (the “slump”).

Technique w-o history_3

Note from the RGDP chart that the real output trend doesn´t change, as it shouldn´t, given that real output is determined by real (supply side) structural factors. What is permanently changed is the rate of inflation (the slope of the price level line). So what the Fed ended up achieving was something desirable and, since it was due for the most part to Fed policy it could conceivably be permanent (as long as the policy of maintaining nominal stability along a level path remained).

The chart also shows what happened when the level path for nominal spending was forsaken. Inflation remains low (because nominal spending growth, its main determinant, remains low) but real output falls to a lower trend level, with the nasty consequences this entails for the level of all its component parts and, therefore, to the level of employment.

The success of the Great Moderation was grounded on two factors: One, on the Fed having attained nominal stability, and two, on having (here luck may have given a helping hand) established the ‘proper’ initial level of nominal spending from which to progress.

Today, instead of realizing we “bucked the nominal spending trend level”, there´s an ongoing “effort” to justify the lower level. The “secular stagnation” hypothesis is the “front runner”. A formal model of it has just been released. Maybe (and hopefully) as Nick Rowe argues in: “On forgetting land in models of secular stagnation“:

So when you go to a helluva lot of trouble to build a model with a negative equilibrium real rate of interest, and it’s a very fancy complicated model, but it totally ignores land, I really wonder where you are coming from. Actually I don’t wonder. I know where you are coming from. You are coming from the town, or the big city, where you can easily forget about land. But even then: you know that stuff your house or condo is built on? That’s called “land” too.

It didn’t used to be this way. The Physiocrats, Malthus, Ricardo, didn’t ignore land. Land was central to their models. Now I know that whether or not you should include something in your model depends on what you want your model for. And that it is perfectly OK to ignore land in some models. But if you have a model of secular stagnation and negative real interest rates that ignores land, well, your model is just hopelessly and fatally wrong. Dead wrong. It’s just not grounded.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.