About Paul Krugman: Having the Right Diagnosis Does Not Mean He Has the Right Cure
This is the second time in less than a month that Paul Krugman’s picture has headed one of my posts. (The other time is here.) That is no accident. Paul is the true monster of the economics blogosphere–as well as in the beleagured redoubts of non-electronic economic journalism that remain. I use the word “monster” in the positive and enviable sense of having a large reach and influence with the words that he writes. (Please, may I some day grow up to be a monster? See the illustration from Where the Wild Things Are.)
In his recent post, “Smuggish Thoughts (Self-Indulgent),” Paul writes this:
I got obsessed with Japan in the 1990s, and I think can fairly claim to have started the whole modern liquidity-trap literature. I approached the Japan problem the way I approach just about all economic problems, building a stylized, minimalist model (big pdf) that seemed to make sense of the available facts and yielded strong conclusions. But does this style of analysis work in the real world?
Well, events provided an acid test. If you believed in the little models I and others were using, you made some very striking predictions about how the world would work post-crisis–predictions that were very much at odds with what other people were saying. You predicted that trillion-dollar deficits would not drive up interest rates; that tripling the monetary base would not be inflationary; that cuts in government spending, rather than helping the economy by increasing confidence, would hurt by depressing demand, with bigger effects than in normal, non-liquidity trap times.
And the people on the other side of these issues weren’t just academics, they were major-league policy makers and famous investors.
And guess what: the models seem to work. It appears that I wasn’t just a successful self-marketer, that I really did and do know something.
Basically, I agree with Paul’s assessments here–his diagnosis of what happened. But I do not agree with his prescription. As near as I can make out (and I am happy to be corrected on this), his number one recommendation has been a large increase in government spending to provide Keynesian stimulus, and his number two recommendation has been for the Fed to promise future inflation above its normal 2% target.
That secondary recommendation I discussed in my earlier post on Paul Krugman. I will not repeat everything I said there, but let me say a few words about the relevant scientific issue. The issue I have with Paul’s analysis there is that he seems to approach the approximate Wallace neutrality that is likely in the real world–which can account for the facts he mentions above–for the perfect Wallace neutrality of his simple model, which would imply that large scale asset purchases by the Fed (as in QE1, QE2, QE3 and Operation Twist) will not work in any direct way, so that the Fed’s only option for stimulating the economy is to promise (or hint at) inflation above 2% in the future.
In relation to Paul’s primary recommendation of a massive increase in government spending in the short run, my main objection is that (assuming we are not willing to contemplate national bankruptcy), every dollar the Federal government ultimately adds to the national debt is a dollar that has to be paid for by taxes further down the road, or by cuts in government spending further down the road that will be hard to bear, given the aging of the population. Except in the case of spending now that can genuinely serve instead of spending in the future, we have to be very concerned about the cost of stimulative spending.
Let me give a simple numerical example to make the point. After the economy gets fully back on its feet, I expect the interest rate to be something like 4% per year in real terms. Suppose we added $2 trillion more to the debt to stimulate the economy and then wanted to keep that extra debt from growing further in real terms so that the growth of GDP could gradually reduce the debt-to-GDP ratio. To do that, we would have to pay the real interest on that extra debt: $2 trillion * 4% per year = $80 billion per year. If GDP by then is a little higher than now, at $16 trillion per year, that is a ½ % addition to the spending to GDP ratio. A lot of the big arguments between Republicans and Democrats are about differences in government spending on the order of about 3% of GDP. So ½ % of GDP difference in government spending due to extra interest payments is actually a very big deal.
So it is a great advantage to simulate the economy by measures that add less to the national debt, the Federal Lines of Credit which I lay out in my second post “Getting the Biggest Bang for the Buck in Fiscal Policy” and have discussed at great length in the other short-run fiscal policy posts on this blog.
(It seems plausible to me that large scale asset purchases by the Fed also have this property of stimulating the economy while adding relatively little to the national debt in the end. I would be glad to see a careful analysis of the likely round-trip financial costs to the Fed of pushing interest rates down and asset values up by buying long-term government bonds and mortgage-backed securities now to stimulate the economy, and pulling interest rates up and asset values down later by selling them to rein the economy in–or alternatively raising interest on excess reserves later.)
It matters how we approach the problems that we face. Paul Krugman deserves a lot of credit for getting the basic diagnosis of our problems right, but he needs to be just as serious about identifying the best possible solutions. Traditional Keynesian remedies or promises of inflation may work to stimulate the economy, but what if there is a better remedy, with fewer undesirable side effects? It is my contention that there is a better remedy, that would have the same effectiveness at lower cost: Federal Lines of Credit. And that is in addition to the possibility that the Fed has already found a better approach in large scale asset purchases, if only it pushes hard enough on its string.
Update: A commenter on Twitter (I’ve lost track of the tweet) points out that the government can stabilize the debt to GDP ratio if it pays only the interest rate minus the growth rate of the economy on the debt each year, rolling over the rest, including rolling over the part of interest payments equal to the growth rate. That makes the long-run picture look less stark than the calculation I make if the interest rate is less than 3% above the growth rate of the economy. For example, approached that way, if the interest rate is only 1.5% above the growth rate of the economy, then the $2 trillion in extra debt would mean a permanent ¼ % of GDP less spending or a permanent ¼% of GDP less taxes.