There are signs of economic recovery. Suppose a year from now it is clear that the economy has indeed recovered. What caused it? Was it: monetary policy; fiscal policy; or did the economy recover by itself?
I don't know the answer to that question; nor does anyone else. If you have one world recovery, and three competing hypotheses to explain that single data point, you don't know the answer. So we resort to theory, and see if theory, backed by multiple data points from past recoveries, can help us answer the question.
Macroeconomic theory tells us how monetary policy is supposed to work; and it tells us how fiscal policy is supposed to work. But macroeconomic theory (at least, the modern, canonical, neo-Wicksellian model) does not tell us how economies are supposed to recover by themselves. Like pre-Darwinian biology, it only works if there is some sort of designer or creator to make it work: in this case the monetary and/or fiscal authorities, rather than a divine being.
Microeconomic theory is fine. We don't need to posit divine intervention to explain why a glut of apples is self-correcting. The demand curve for apples slopes down, and the supply curve for apples slopes up. So a glut of apples causes a fall in the price of apples relative to other goods, which causes an increase in quantity demanded, a decrease in quantity supplied, and the glut cures itself.
Microeconomics doesn't say that the market for apples works perfectly, or instantly; any more than a Darwinian would say that natural selection works perfectly or instantly, or that artificial selection cannot sometimes improve on natural selection. But it does work, and we understand why it works. Microeconomics is not pre-Darwinian (except in the sense that Adam Smith preceded Charles Darwin).
It's the most fundamental question of macroeconomics: is the economy as a whole self-correcting in the way that individual markets are self-correcting? Does a general glut of all goods tend to cure itself in the way that a glut of apples tends to cure itself?
And it really bugs me (and it ought to bug me) that I don't have a good answer to that question (and nor does anyone else).
A fall in the price of apples relative to other goods will cause producers to substitute away from producing apples to producing more other goods instead. That's why the supply curve of apples slopes up. But that can't be why the aggregate supply curve slopes up, because we are talking about all prices falling when there is a general glut. Macroeconomic theory says that the Long Run Aggregate Supply curve is vertical. And indeed it should be vertical, if workers still want to work, and technology and other resources haven't disappeared in a general glut. That's what we mean by "self-correcting" anyway: eliminating a general glut so we get back to something that looks at least vaguely like how an economy ought to be operating, even if you don't like the phrase "full employment".
A fall in the price of apples relative to other goods will cause consumers to substitute away from consuming other goods to consuming more apples instead. That's why the demand curve for apples slopes down. But that can't be why the aggregate demand curve slopes down, because we are talking about all prices falling when there is a general glut.
Macroeconomic textbooks often draw a downward-sloping Aggregate Demand curve, but the explanations for that downward slope are less than convincing. In an open economy with fixed exchange rates the AD curve slopes down for the very same reason that the demand curve for apples slopes down (just substitute "Canadian-produced goods" for "apples"); but that won't work for a global general glut. In a closed economy (like the world) the only valid explanation for a downward-sloping AD curve is that a fall in the price level will cause an increase in the real money supply, if the nominal money supply is exogenous (and the increase in the real money supply will cause people to attempt to substitute away from holding money towards, directly or indirectly, buying more goods). Even then, we worry that falling prices might cause expected deflation and increase the real demand for money by more than the real supply of money has increased, leading to an upward-sloping AD curve.
And to show how little we are convinced by that explanation of a downward-sloping AD curve, we ignore it altogether in the Neo-Wicksellian models that are the modern workhorse of applied macroeconomic theory. The central bank sets a nominal interest rate in those models; the nominal quantity of money, if it even appears in the model, is an endogenous irrelevancy.
In fact, modern neo-Wicksellian models are exactly the same in this regard as the very simple old-fashioned Income-Expenditure "Keynesian Cross" model I first learned in school. They have zero automatic tendency towards anything that might have anything to do with "full-employment". They don't self-correct; general gluts do not cure themselves, ever, except by sheer chance, or intervention by monetary and/or fiscal designers. (Actually, neo-Wicksellian models have a tendency to automatically move away from anything resembling a full-employment equilibrium with stable prices: deflation begets expected deflation, rising real interest rates, and increased deflation; inflation begets expected inflation, falling real interest rates, and increased inflation.)
Well, you might say, perhaps those models are nevertheless right, and that's just how the world is. There is no self-correcting force at the macroeconomic level. General gluts do not cure themselves.
I distinctly remember myself at age 17 being unsatisfied with that answer when first introduced to macroeconomics and the Keynesian Cross. I was like a pre-Darwinian atheist confronted with the Argument by Design. "That can't be right" I thought, "because Keynesian policy is a very recent invention, and if this model were true it would be a miracle if the economy worked at all, and chances are that we would have relapsed back into the Stone Age sometime over the last few centuries".
I feel the same way now, only more strongly, because the Neo-Wicksellian model doesn't just have no tendency to self-equilibrate; it has a tendency to self-disequilibrate. The Anthropic Principle rules it out. We haven't always had any sort of Keynesian monetary and fiscal policy, let alone good monetary and fiscal policy. If there were no self-correcting macroeconomic process, I wouldn't be sitting in front of a computer now and writing this blog post. I would be knapping flints, or something.