So how much we spend on supporting the economy in 2010 and 2011 is almost irrelevant to the fundamental budget picture. Why, then, are Very Serious People demanding immediate fiscal austerity?I am in the camp of people who think it would not be prudent to increase government debt too much higher than it is at present in the US in particular, except in a worse emergency than we currently face. Part of my issue is general prudence - once in a while, serious country-threatening shit happens (wars, financial crisis, etc) and a government should maintain some margin to handle these, and not expend all its powder to promote growth or reduce unemployment. And in particular, I think the U.S. faces some serious long-term issues that are only just beginning to bite, and it's not prudent to max out the credit card ahead of them.
The answer is, to reassure the markets — because the markets supposedly won’t believe in the willingness of governments to engage in long-run fiscal reform unless they inflict pointless pain right now. To repeat: the whole argument rests on the presumption that markets will turn on us unless we demonstrate a willingness to suffer, even though that suffering serves no purpose.
And the basis for this belief that this is what markets demand is … well, actually there’s no sign that markets are demanding any such thing. There’s Greece — but the Greek situation is very different from that of the US or the UK. And at the moment everyone except the overvalued euro-periphery nations is able to borrow at very low interest rates.
So wise policy, as defined by the G20 and like-minded others, consists of destroying economic recovery in order to satisfy hypothetical irrational demands from the markets — demands that economies suffer pointless pain to show their determination, demands that markets aren’t actually making, but which serious people, in their wisdom, believe that the markets will make one of these days.
However, I also think there is something fundamental to the nature of debt that Krugman seems to me to be missing, which explains why the fact that interest rates are currently low is not as reassuring as he thinks, and why it's reasonable to worry about "hypothetical irrational demands from the markets" as he puts it. I would like to make my own mental model more explicit by presenting a simple model of investor behavior under default risk. I'll present it qualitatively, but it should be fairly clear how it could be made at least loosely quantitative.
The basic issue I see is that the collective confidence of investors is a key variable here. If investors are all comfortable that a government can service its debt, they demand low interest rates. In this situation, the cost of servicing the debt is moderate, and the government has no difficulty doing so. However, if investors lose confidence, then they demand high interest rates. This causes the cost of servicing the debt to go up, and makes it harder for the government to do so, and makes default more likely. This can cause a runaway loss of confidence, until interest rates are at predatory levels (high enough that investors hope to recoup their money in the remaining time before default, and/or via a negotiated settlement in a partial default).
Thus it seems to me there is a dual equilibrium situation for countries with significant debt. In the low equilibrium, interest rates are low, and the government can service the debt. In the high interest rate equilibrium, interest rates are high and the government will struggle to service the debt. In either equilibrium, investor perceptions will be correct.
I illustrate this with a diagram that looks like this:
Here the X-axis is interest rates, and the Y-axis is something like the difficulty of the overall economic system being in this state. The little black circle represents the current state of the system, and the little arrows to the left or right represent the range of currently likely fluctuations. You can think of the blue curve as an upside down probability distribution, or you can think of it via a physical analogy as a (gently vibrating) surface on which the little black ball is rolling around.
The hump between the two equilibria arises from the social agreement between investors - if most investors have confidence, then the system is in the low equilibrium, if most have lost confidence then it's in the high equilibrium, and the tendency of investors to herd together causes the in-between states to be less likely.
(Stray technical note to anyone passing with relevant knowledge - yes, this is basically the phi^4 model of statistical mechanics).
If you think of a low debt country, where it's very implausible that the country couldn't pay its debts even at interest rates much higher than at present, the diagram would look like this:
The high-interest-rate equilibrium barely exists, and there is essentially no chance of the system getting into it. Now, if the country runs deficits for a long time, so that the debt level increases, it starts to be somewhat more possible for the high-interest-rate equilibrium to exist. However, because there is a barrier between the two equilibria, and the investors started in the low equilibria, they stay there:
If you just look at market interest rates here, things continue to look ok - interest rates are a little higher, and maybe fluctuate a little more, but the situation as far as observable variables go looks fine. However, if we could really observe the whole blue curve, we could see the situation was increasingly dangerous. As debt increases further, it starts to get really dangerous:
Finally, the low equilibrium becomes sufficiently poorly defined that some random event can cause us to jump out of it and run down-hill into the high-interest-rate equilibrium:
This is the run-on-the-government phenomenon. So the characteristics of this model are that interest rates only respond relatively tepidly to increasing debt up to some threshold, beyond which they abruptly run away to much higher values. There is a highly non-linear response built into the system.
It appears to me that something rather like this just happened to Greece (graph from here):
Greek debt levels were increasing steadily for a long time (decades), but investors rather abruptly lost confidence collectively and started demanding much higher interest rates to loan to Greece (until the EU responded and, at least temporarily, reassured the market).
So I think a lot of the angst in the US and the UK about the debt level comes from an intuitive feeling that perhaps we are in this situation:
We are not in immediate trouble, but some unknown-but-not-too-large additional increment to the debt will push us over the edge.
Now, it's hard to assess what the unobservable parts of the curve really look like, so it's hard to say for sure. But, the point of this kind of model is that merely noting that long-term interest rates are currently low does not provide much reassurance that they won't rise dramatically in future.