I think it was Ronald Reagan who once quipped that an economist is someone who sees something working in practice and then asks whether it might also work in theory. (A good one, I have to admit).
Many can relate to this sentiment. It is based on the idea that any given body of data largely "speaks for itself." We do not need abstract theories to interpret the world and help guide our decision making. Just look at the evidence, and use some common sense.
Well, one problem with this is that "common sense" frequently means different things to different people (e.g., my common sense; not yours, you dope). If it is indeed true that the data speaks for itself, how is it possible for people not to agree on what the data is saying? Evidently, data does not always "speak for itself;" it is subject to interpretation. And when data is limited, it is frequently difficult to discriminate between competing interpretations (theories). Scientists struggle to find the most plausible interpretations--a set of tentatively-held hypotheses, destined for modification (or even destruction) as the science progresses.
Why am I saying this? Well, in part because Krugman seems to toss a bone to this general idea in the opening paragraph of his article:
One thing that's been conspicuously missing from the back-and-forth among economists about fiscal stimulus is anything resembling a fully specified model. To some extent that's O.K. [...] But there are dangers in relying entirely on implicit theorizing: you can find yourself saying things you think must be true, but that turn out not to be true even in a simple model with maximizing agents.
What's this? It is a rather odd statement coming from a person who generally speaks as if he knows "the truth" on any given matter. But here (and recently elsewhere), he laments the absence of anything resembling a "fully specified" (read: "fully coherent") model. What's this all about?
Perhaps it is an implicit admission on his part that the historical evidence (as of Dec. 29, 2008) does not "speak for itself" --at least, not in a manner that would support his long-held and steadfast belief on this matter. And so, if the Conscience of a Liberal can finally admit to this, I should hope that others might now feel free to follow in his example.
Alright then, on to the main point of his article. His goal, quite clearly, is to demonstrate that what he strongly believes to be true actually possesses some theoretical legitimacy. He chooses as his framework of analysis a more-or-less standard New Keynesian (NK) model. The NK model traditionally downplays financial market frictions and highlights, instead, frictions in nominal wage/price adjustments. I say this to alert the non-specialist to the fact that Krugman wants us to understand the current liquidity trap as a phenomenon that arises in a world where financial market imperfections play no central role. (This may or may not be a legitimate abstraction for the purpose at hand--I admit to having my doubts).
On his blog he warns us that the paper is likely to be incomprehensible (because it was written in a hurry). Well, it is not so much incomprehensible as it is incomplete. It is, as he admits, simply a sketch of an argument.
And his argument is roughly this. Imagine that the economy is hit by a "shock" that renders households temporarily more patient; i.e., they now wish to delay their consumption purchases. This generates a large decline in consumer demand--on the flip side, a "savings glut." (Note: in the recent recession, consumption remained relatively stable--it was investment spending that tanked). The economy "wants" a lower interest rate and normally, this desire is accommodated by the Fed. But when the zero lower bound is reached, the lack of nominal price adjustment implies that current GDP (income) must decline to frustrate the glut in desired saving. The decline in GDP could be averted, or at least mitigated, by an appropriation of resources from households (in the form of a lump-sum tax), which the government is required to spend (consume).
Maybe there is something to this--I don't know. The result appears to hinge critically on the assumption of sticky prices, and one might legitimately wonder whether there is enough "stickiness" out there to render the effect quantitatively large and persistent. The interested reader may wish to consult: When is the Government Spending Multiplier Large? (Christiano, Eichenbaum and Rebelo, 2009). Their calibration appears to have an implausible degree of price ridigity built in, but please correct me if I am wrong.
I would like to say a few things about the way the K-man concludes his analysis:
The bottom line is that while we usually think of Keynesianism as the preserve of ad hoc models, in this case doing it "right"--using a macromodel with maximizing agents and a proper concern for intertemporal constraints--actually suggests a very strong case for big government spending in the face of a liquidity trap [...] So let's get those projects going.
Hmm...a case for big government spending? That's quite the conclusion from a qualitative model. Oh well, just a minor slip of the tongue, I'm sure.
Second, I do not think of Keynesianism as the preserve of ad hoc models (even if Krugman may at times be). I think that JM Keynes was a very careful theorist, especially given the tools at his disposal. And I would like to stress that Keynes was not ashamed of theory:
But [this book's] main purpose is to deal with difficult questions of theory, and only in the second place with the applications of this theory to practice." - John Maynard Keynes, The General Theory of Employment, Interest, and Money, p. v.Finally, Krugman leaves the impression that he has proven something very robust; a property that would hold in any NK model. A recent paper suggests that this is not the case: Fiscal Policy in an Expectations Driven Liquidity Trap (Mertens and Ravn, 2010). Here is the abstract:
Apparently, there is still much to learn...even if your initials are PK.
We examine the impact of fiscal policy interventions in an environment where the short term nominal interest rate is at the zero bound. In the basic New Keynesian model in which the monetary authority operates a Taylor rule, globally multiple equilibria arise, some of which display all the features of a liquidity trap. A loss in confidence can set the economy on a deflationary path that eventually prevents the monetary authority from adjusting the interest rate and can lead to potentially very large output drops. Contrary to a line of recent papers, we find that demand stimulating policies become less effective in a liquidity trap than in normal
circumstances. The key reason is that demand stimulus leads agents to believe that things are even worse than they thought. In contrast, supply side policies, such as cuts in labor income taxes, lead to relative optimism and become more powerful.