I kept telling myself that I didn’t have a problem, that I could stop reading the comments in this post whenever I wanted. Well I think I need to go cold turkey. So let me make three main concluding points:
POINT #1: The accounting identities don’t prove the things that the MMTers think they prove, at least not by themselves. They must be supplemented with a particular macroeconomic theory. Rival theories, which yield the opposite conclusions, are just as consistent with the accounting identities.
That was of course the main point of my Mises article, but in the comments at the blog discussion, “MamMoTh” unwittingly illustrated it beautifully. Commenter Eli was arguing with Mammoth, saying that regardless of the government’s behavior, the private sector could save more if it just reduced its consumption.
Mammoth disagreed, saying: “Any individual can adjust his/her saving/consumption ratio. But the whole private sector can reduce consumption, which will reduce income but leave savings unchanged.”
To drive home the point with math, rather than mere words, Mammoth then wrote:
Y = C+I+G
So consumption does not affect aggregate savings.
But in Eli’s framework, when consumption goes down, investment goes up, and Y stays the same. That is totally consistent with the first equation above. Mammoth looks at Y = C+I+G and thinks a drop in consumption is handled by a drop in Y, but it could just as well be handled by an increase in I.
And so then in the reduced form, where S = I + (G-T), when I goes up, savings can go up on the left hand side to keep the equation balanced.
POINT #2: Neither the Austrians nor the Keynesians deny that the Fed could print whatever amount of money is necessary, in order to avoid a technical default on government bonds. We are saying that it’s possible this would cause large price inflation, and that’s why it’s dangerous for the government to run huge deficits year after year.
In several places, the MMTers took umbrage when I (repeating Krugman) said that they believed “deficits don’t matter.” They came back and said yes government budget deficits do matter, but not because of “solvency” issues, rather because of the purchasing power of the USD.
If that’s really the argument, then we can all shake hands and go home. Krugman acknowledges that in his critique of MMT:
[O]nce we’re no longer in a liquidity trap, running large deficits without access to bond markets is a recipe for very high inflation, perhaps even hyperinflation. And no amount of talk about actual financial flows, about who buys what from whom, can make that point disappear: if you’re going to finance deficits by creating monetary base, someone has to be persuaded to hold the additional base.
If you don’t believe me, I encourage you to go back and re-read his post. Krugman isn’t saying, “Deficits eventually matter, once we get out of the liquidity trap, because Uncle Sam might default if he gets in over his head.” No, Krugman specifically concedes that the Fed could just monetize all the debt. His point is, that could lead to runaway price inflation.
Now maybe Krugman is wrong for thinking that, but you don’t advance the argument by saying, “A government issuing its own currency can’t go insolvent.” Nobody is denying that.
POINT #3: Even in its textbook version, “crowding out” is consistent with higher private sector saving. What is being crowded out is private investment.
In a completely unrealistic, everybody-is-a-rational-robot Chicago School model, when the government [UPDATE: A few minor edits because it matters whether the government deficit is due to a tax cut versus a spending increase…]
doesn’t raise taxes and yet increases deficit spending maintains its current spending level but gives a deficit-financed lump sum tax rebate, that will cause taxpayers to save more, in order to pay for the higher taxes down the road needed to finance the additional debt. OK, so MMTers aren’t at all scoring points against the “crowding out” hypothesis by showing that private sector savings rates went down during the Clinton surplus years, and up during our recent period of massive government deficits.
Now if we relax the perfect Ricardian equivalence, a more real-world scenario in which “crowding out” occurs, goes like this: The government runs a deficit and thus increases the demand for loanable funds. This pushes up interest rates higher than they otherwise would be. People expect higher taxes, and so they save more, which partially offsets the interest rate hike. However, the effect isn’t perfect, and on net interest rates are higher than they otherwise would be. Therefore private business invests less. In effect, the government has diverted some of the private sector savings that would have gone into private investment, and instead it is being spent as G, not I.
The data are totally consistent with that interpretation. The following chart shows Gross Private Domestic Investment as a share of GDP (top line), versus “net government saving” as a share of GDP (bottom line). Note how they move in close tandem. When the government deficit shrinks, private sector investment goes up. When the government deficit balloons, private sector investment collapses.
So look, I can use the same sectoral balance equations, and I can point to empirical data going back decades, which incorporates periods of government surpluses and massive deficits. The MMT crowd should stop thinking they have a monopoly on accounting truisms and historical evidence.