Hysteresis

…is a controversial concept, and the answer to the question of whether or not it really is a “thing” will have implications in the ongoing competition between “Classical” and “Keynesian” type theories. (More on that later, of course.) For this first, simple explanation, think of “memory foam.” The question is: does the economy ‘forget’ a downturn after it bounces back to its original path of growth (classical equilibrium model), or does it retain a more or less permanent impression–a new, lower path–going forward (hysteresis, not necessarily accepted by all “New Keynesians” at this point).

The policy tradeoff between the unemployment rate and the rate of inflation (expressed by the “Phillips Curve”) is widely accepted by the economics profession. There are a variety of mechanisms invoked for this relationship, but they result in virtually the same model, where a policy that lowers inflation leads to higher unemployment. Here is one easy intuition: imagine a factory owner is facing a higher sales price for her product (due perhaps to increased demand in the marketplace). She will want to ramp up production, which in the short run will require more workers. (Investments in machinery–including robots–are long-run, not short-run, factors.) As the labor market tightens, it may become necessary to offer more money (wage inflation). This feeds the inflationary pressure on the general price level while lowering the unemployment rate. Now imagine the exact opposite process when demand is dropping: things go on sale (sometimes permanently) while the demand for labor drops. This is a period of “disinflation”, where wages and prices drop while economic output shrinks.

Notice how I have described a microeconomic mechanism while (with a slight of hand) extending it to the entire economy. This is called “giving macroeconomics a micro foundation.” There are many ways to go about this, since economics suffers from the problem (familiar to all scientists) that the data are underdetermined by theory. Which is to say, there are any number of theories that could explain the same result. An alternative theory explains inflation and unemployment rates through “imperfect information” about future price levels, but gets to the same result.

Let’s not play dumb here. It is easy to imagine that a short run variation in demand can lead to quick adjustment to the labor force under normal conditions, and therefore a lot of resilience in supply. We see it happening when a business gives its part-time staff more hours to work when a busy season is underway, then cuts back when things slow down again. But everything could change when the slowdown is deep and long lasting. Imagine the business has to close “temporarily.” People will look for other jobs and may not be available when it is time to re-open. When industries lay off hundreds of thousands of workers for several years at a time, it is unlikely that those same workers (along with their skills) will be readily available if and when the economy gets better.

The classical (non-hysteresis) theory depends on the “natural-rate hypothesis,” where the economy taken as a whole has a certain “potential” at any given moment, a sort of maximum output level where all resources (especially human) are fully utilized. Even at this level of output there will be a “non-accelerating inflation rate of unemployment” (NAIRU), because some people are always between jobs for a variety of reasons. However, the rate of unemployment is not as simple a thing as one might think, since it is a ratio of the number of people who haven’t yet found a job to the number who want a job at any given point in time. It turns out that the workforce participation rate (against which the unemployment rate is calculated) actually changes all the time.

We will see that Larry Summers presents data from the past few decades to show that there do seem to be more or less permanent changes to the workforce with every economic downturn. I will do some of my own drilling into the vagaries of measuring unemployment as well.

Next time: Secular Stagnation

We are continuing with Larry Summers’ talk at the N.A.B.E. in early 2014 entitled: “U.S. Economic Prospects: Secular Stagnation, Hysteresis, and the Zero Lower Bound.” As gibberishical as that sounds, it is less so than a lot of academic paper titles. Nevertheless, many of my readers will be seeing familiar words used in a new way here, perhaps wondering, “Zero lower bound of what?” and “Hysteresis sounds Freudian,” and even, “Is the earth’s rotation slowing?” I will give a simple definition of each of these concepts in order of increasing complexity, which puts them in the reverse order from Summers’ title. In future posts we’ll explore them at some depth.

Zero Lower Bound:

This refers to interest rates, which must always be above zero if they are to have any meaning. You loan me $100 at 6% interest, and I pay you back $106. A 0% interest rate would mean I pay you back exactly $100. A negative (-6%) interest rate would be when we agree that I will pay you back only $94. The last one doesn’t make any sense: we never see loans with “negative” interest rates.

But actually–when there is inflation–it is possible for an interest rate to be negative in inflation-adjusted terms. Let’s say the rate of inflation is 2% and you loan me $100 at a rate of 1%. Fast-forwarding a year, when I pay you $101 it is worth slightly less now (when adjusted for inflation) than the $100 dollars was worth when I borrowed it. On the other hand, it is worth slightly more than the $100-dollar bill that you stuffed under your mattress, which is now worth only the equivalent of $98.04 in last year’s dollars. By comparison, the $101 I just gave you is worth $99.02. 2% inflation over the past year means that today’s $102 is worth exactly the same as $100 was worth a year ago. To stay even with inflation, you would have had to charge me a nominal interest rate of 2% to match the expected rate of inflation. In that case, the real interest rate would have been 0%. In the case where inflation was 2% and the loan was at 1%, there was a negative real interest rate of 1% – 2% = -1% (nominal interest rate minus inflation rate equals real interest rate). So sometimes interest rates are negative when inflation is taken into account. (You may have figured out by now that for economists, ‘real’ is shorthand for ‘adjusted for inflation’.)

What professor Summers is talking about is a key interest rate for the economy which is set by the Federal Reserve, the “Fed Funds Rate,” which is what banks charge each other for overnight deposits. The Fed can’t control this rate directly–there is no chalkboard at the main entrance with today’s special rate posted. Instead, they influence the rate by buying and selling securities. When they buy securities from the open market, there is more cash “out there.” When they sell them, they are sucking up cash from the market. This adjustment to the money supply leads to changes in the interest rate according to the forces of supply and demand. They set a “target rate” and buy and sell securities until the going rate is close enough to the target. (As an example of how indirect this control is, the Fed Funds Target Rate has been 0.25% for quite a while, but the effective market rate is only 0.13%–they seem to be having trouble getting it up.)

Because banks always have the option of just sitting on cash rather than lending it, it is impossible to get the rate to go below zero. This is a problem for the Fed when the economy is in recession and they want to cut interest rates to stimulate investment. Once they hit the zero lower bound, there is no place else to go.

OK, I am out of time for today, but tomorrow I will take up hysteresis, and probably get to secular stagnation in the post after that.

I don’t imagine that the majority of my readers have degrees in economics. Even having one myself (albeit a mere bachelor’s) I am often befuddled by the jargon and mathiness of professional academic papers. I even find myself googling terms and references just reading Bloomberg, the Financial Times, or The Economist. My goal, then, is to fill the gap between the professional economists and my readers who, while intelligent and curious, have other specialties.

So let’s begin with the title of professor Summers’ talk: “U.S. Economic Prospects: Secular Stagnation, Hysteresis, and the Zero Lower Bound.” I have run these terms by several people and have confirmed that they are meaningless to the average person. I said in my previous post that I would do my best to translate, so here goes.

The general topic is, of course, U.S. Economic Prospects. Dr. Summers has been invited to give a speech to a group of business economists — as opposed to academics. This means they work as economists for companies (think ‘applied math versus pure math’), and as such are always interested in what the current data actually mean for the immediate future. Summers emphasizes his understanding and appreciation of this distinction in the introduction of his speech, saying

Indeed, I think it is fair to say that some of the themes that are today central to discussions of academic macroeconomists, but that had receded from the debate for many years, were always kept alive at the National Association for Business Economics.

Another way of understanding this is that in the world of business, it is more important to be right in a real-world sense than to take credit for having the latest, greatest theory. Which takes us to the bigger question of “what is macroeconomics for?” Summers explains that the world of macroeconomics today is different in substantial ways from that of just a few years ago, thanks to the major disruption to financial markets that began in 2007. Prior to the crash, macroeconomists were convinced that the days of large-scale depressions were in the past, that during recent decades a “great moderation” had taken hold. As if, “we know too much to fall into the errors of the past: our sophisticated models give our central bank the power to use minor tweaks to keep the economy on a steady upward path of growth.” No longer would we be victims of wild swings in the business cycle.

The Great Recession gave the lie to that notion, and rudely at that. So Summers sees the opportunity to question everything: to do meta-macro (as a philosopher might say), or, if you like, macro-macro, where even the business cycle itself comes under interrogation: “As I shall discuss, there is room for doubt about whether the cycle actually cycles.” He doesn’t need to point out to the economists in the room that the “recovery” since 2009 (when the recession technically ended) has been anemic by most measures. The big question he wishes to pose is whether the very slow growth we have seen in the past five years is symptomatic of a gradual rebound from a severe low in the business cycle, or if it is indicative of a “new normal,” a much flatter growth curve going forward. For the latter to be the case there would have to be structural reasons for it, such as shifts in demographics or technology that make the prospects for future economic growth permanently and broadly diminished — secular stagnation.

In my next post I will define the three terms in the speech’s title: secular stagnation, hysteresis, and the zero lower bound, after which we will have to drill into each of them at some depth.