A witty geometrician might observe that the Economist has overturned Euclid, in drawing this familiar graph: instead of two points determining a line, two lines determine a point. The joke is actually profoundly important: this economic analysis is overdetermined. Whatever price and quantity transacted, there are always two available explanations: supply and demand.
What one can actually observe is the price and quantity transacted. The rest of the supply or demand curve is a counterfactual supposition: an act of imagination tempered by subjective assessments of plausibility.
The result is arguments between economists, over whose counterfactual is better.
Here's Brad DeLong:
Consider the claims – rampant nowadays in the US – that further government attempts to alleviate unemployment will fail, because America’s current high unemployment is “structural”: a failure of economic calculation has left the country with the wrong productive resources to satisfy household and business demand. The problem, advocates of this view claim, is a shortage of productive supply rather than a shortage of aggregate demand.Now, I am sympathetic to Professor DeLong's political viewpoint, so let it be understood that my criticism, here, is of his economics, and not so much his politics.
But it should be easy . . . to tell whether a fall in sales is due to a shortage of supply or a shortage of demand. If a fall in sales is due to a shortage of demand while there is ample supply, then, as quantities fall relative to trend, prices will fall as well. If, on the other hand, the fall in sales is due to a shortage of supply while there is ample demand, then prices will rise as quantities fall. . . .
Or consider the claims – also rampant these days – that further government attempts to increase demand, whether through monetary policy to alleviate a liquidity squeeze, banking policy to increase risk tolerance, or fiscal policy to provide a much-needed savings vehicle, will similarly fail. Once again, . . . ask whether the economic problems that current levels of government debt are causing reflect too much public debt supplied by governments or too much public debt demanded by the private sector. If the problem were that supply is too great, then new emissions of government debt would be accompanied by low prices – that is, by high interest rates. If the problem were that demand is too great, then new emissions of government debt would be accompanied by high prices – that is, by low interest rates..
This is a very common form of analysis for economists. Marshall's Cross is iconic for a reason. There's also a bit semantic generalization going on, a bit of intellectual slipping and sliding away from the intellectual foundation. Professor DeLong has moved from Marshall's analysis of a single market, to arguably, an aggregation of markets. That entails some very big problems, which he doesn't bother with. And, I won't bother with them, either, for the moment.
What interests me, at the moment, is that he is so confident that Supply AND Demand can be treated as a problem of Supply OR Demand. The basic theory is the former. Brad's theory is the latter.
It's a slippery form of argument. You need two points to determine a line, and Brad gets his two points from change in time: his points are before and after. Of course, that only gets you one line, not the two of supply and demand. So, he argues that his two points must describe either supply or demand, depending, as he would have it, on circumstance.
There's a lot of imagined stuff, here, and very little observed fact, and what is observed is not enough to distinguish cases. Brad is simply wrong when he asserts "it should be easy . . . to tell whether a fall in sales is due to a shortage of supply or a shortage of demand". According to the theory, which, by the way, doesn't define anything as a "shortage of supply" or "shortage of demand", it should be impossible to tell attribute a price/quantity entirely to demand or supply, because the price/quantity outcome is jointly determined.
The supposition that supply slopes upward, increasing with price, and demand slopes downward, decreasing with price, is just that: supposition. It rests on some auxiliary hypotheses -- the famous law of diminishing returns and the like -- that express our intuition and sense of what is plausible. There's no more foundation for them, really, than the plausible notion that heavy objects will fall faster than lighter ones, propelled by the greater force evidenced in their greater weight.