Scenario for a meltdown

I recently posted a challenge on a listserv on teaching economics, which essentially asked people to use our current understanding of the macro economy to make sense of what has happened over the past year. I wanted to comment here about the issues.

Suppose it’s June 30, 2008. What do we know about the current state of the economy? (1) Employment has been falling (by small amounts) for about six months. (2) Housing values have been declining, slowly in some parts of the country, severely in other parts of the country, for nearly two years, so the residential housing part of the construction business is in serious collapse. (3) The rate of inflation (as measured by the CPI) has increased (slowly, and in fits and starts) from about 1.5% (using year-over-year percentage changes in the monthly CPI-U) since early 2004 to over 5% by the middle of 2008. (4) Real GDP has been growing at (generally) between 2% and 3% on an annual-rate basis, but growth has begun, in the second quarter of 2008, to slow.

Now suppose we add to this one piece of information. Over the next nine months, the money supply, measured as M1, will grow at a 15.4% annual rate. What would we expect to happen?

I suspect we’d get a clear majority answer from most economists, including those who pay more attention to macro—inflation will continue to accelerate. I suspect there would also be a significant minority that said, “Why did the Federal Reserve allow that kind of growth in M1? Something bad must have happened to the economy,” and some effort to come up with what that might have been.

So let’s add a second piece of information. In addition to the accelerated (15.4% annual rate) growth in the money supply, we also know that the rate of inflation fell abruptly and sharply, until by March 2009, the annualized rate of inflation had declined to -0.5%--the general level of prices was actually falling. Now what would we get?

Well, one thing that my post did get as a response was, “Obviously, the velocity of money has declined. Well, purely by definition that almost had to be true:

Mv = PQ

is the quantity equation. If growth in M (money) accelerates and if the rate of inflation decelerates, and if there’s no change in the growth in output, then v (velocity) has to fall. But that’s boring. The question is, what’s happening in the economy that is causing velocity to fall?

And the question really is, what could happen that would cause velocity to fall, if you have to answer this question before-the-fact, rather than after we have gone through the decline in velocity.

Again, I suspect there would be a majority answer, which would go something like this: Suppose something happened that raised the riskiness of most financial instruments. (We’ll worry about what this could be later.) Note that this increase in risk has to affect the US economy directly. How will people react? They will move their financial assets to cash or to risk-free cash-equivalents (short-term Treasury bills). People will acquire cash, not to spend, but to hold as an asset. (We can express this by saying that the demand for money as a store of value not as a medium of exchange—or, if you’d like it in Keynesian terms, that the demand for money for precautionary reasons--has increased.) One effect of this will be to drive down the interest rate in risk-free cash equivalents. A second effect will be to reduce the velocity of money.

Now the Federal Reserve, observing this, and knowing that the likely outcome is reduced spending (the aggregate demand curve shifts to the left (or, in older Keynesian terms, the IS curve shifts to the left) and a decline in output. The Fed responds in typical fashion, adding to bank reserves in order to encourage spending. But if the demand for cash (or cash-equivalents) is extremely interest elastic, people and institutions will hoard cash—we are in a liquidity trap.

The tricky part is specifying what might cause the increase in the perceived riskiness of many (most?) financial assets. I suspect there might be many candidates, and I suspect some people would have pointed to mortgage-backed securities as one candidate.

My point here is that we can tell a consistent story, using fairly standard (indeed, entirely Keynesian) macro theory. It’s easier after-the-fact, but I think it’s also clearly possible to construct a reasonably accurate scenario before-the-fact as well.