By Karl Smith
Naturally there was pearl clutching all around after Henry Blodget posted that Niall Ferguson claimed John Maynard Keynes was a short-sighted spendthrift because he was gay and had no children.
I can let other people deal with the bigotry-social impropriety connection. I want to address to issues of what I see as just blatant wrongness.
The first is the suggestion that Keynes and Keynesians are short-sighted. This is so plainly false as to embarrass everyone involved. Keynes was intensely interested in the future and engaged in what we now term futurism. Trying to build and at least test for consistency rigorous scenarios about the future of humanity. His piece, The Economic Possibilities of Our Grandchildren is famous.
The face-saving excuse here is that Keynes seemed to display a low discount value: in the long run we are all dead, he famously wrote. The sentiment may strike some people poorly, but it is one I share deeply and I can say it doesn't arise from trying to ignore future consequences but from taking them extremely seriously. Keynes seems to have have a similar experience, as he wrote the following, which is deeply after my own heart.
The outstanding fact is the extreme precariousness of the basis of knowledge on which our estimates of prospective yield have to be made. Our knowledge of the factors which will govern the yield of an investment some years hence is usually very slight and often negligible. If we speak frankly, we have to admit that our basis of knowledge for estimating the yield ten years hence of a railway, a copper mine, a textile factory, the goodwill of a patent medicine, an Atlantic liner, a building in the City of London amounts to little and sometimes to nothing; or even five years hence. In fact, those who seriously attempt to make any such estimate are often so much in the minority that their behavior does not govern the market.
When I said much the same to Betsy Stevenson, on Up With Chris, she remarked that she knows her kids will need good schools 20 years from now. I am not so sure.
Thinking deeply about the future is an exercise in realizing how much you don't know. You are troubled as to how you are ever going to make good long-term decisions, concerning investments in family, career, financial instruments, etc.
Then the horror sets in when you realize that even as utterly ignorant as you are, you know more and have thought about this issue more deeply than just about anyone in a position of power in politics or business. If you don't know anything, then they definitely don't know anything and they don't even seem to know that they don't know anything.
We are doomed.
Luckily for me I was born after F.A. Hayek and so I can take refuge in the region that lies between instinct and reason. In Hayek's words, tradition. And, I can know that the forces of natural selection will act on tradition to make everything more or less, kinda-sorta work out, sometimes. Though things can get quite bumpy along the way and of course all stories end exactly the same way: Everyone you care about dies. Every thing you care about erodes away.
On the bright side, crushing existential uncertainty and inevitable doom reminds us that if we can just make sure the world doesn't end tomorrow, we are doing well. And, from time-to-time the world ending tomorrow is a serious possibility, so there is useful work to be done here. Keynes would like to see that we do it.
So, this is not insouciance. It's the realization that cataclysms are what happens when you are
making other plans at cocktail parties telling bullshit stories about how you are setting things up for the long run.
The second, far shallower point is this. Henry Blodget writes:
Second, Keynes's policies did not suggest that he did not care about future generations. On the contrary… For the sake of both future generations and current generations, Keynes believed that governments should run deficits during recessions and then run surpluses during economic booms. Politicians have never seemed to be able to follow the second part of Keynes' prescription-they tend to run deficits at all times-but it seems unfair to blame this latter failing on Keynes.
Fair or not, it is simply untrue. Governments can and do run absolute surpluses during booms. Surely, this is not the first time the blogosphere has seen this graph.
A seemingly small positive blip - we'll get to that - but it came exactly when a Keynesian would have wanted it to. Before, we get into the blip check out state and local government savings.
Exactly the trend you would expect. Though, the magnitudes are small compared to the relatively large importance of state and local governments in the the United States. State and locals provide most of the actual government services.
The U.S. Federal government mainly writes checks. However, in that role it is very special. Many firey blog posts have been written trying to get at exactly why the U.S. Federal Government is so different, but here is the key fact.
In blue is the Federal Deficit as a percentage of GDP. In red are Federal Interest Payments on the Debt as a percentage of GDP. Its not clear that there is any reliable pattern, but if anything, the tendency is for these two to move in opposite directions. We can invert to check.
That looks like a somewhat - though not totally - better match. Which would imply that as the federal government tries to borrow more, it pays less in total interest. Not just interest rate, total interest.
No let's do the same using total debt to GDP
This is more interesting because it looks as if the pattern held, right up until recently. More debt meant more interest payments. Then the relationship broke down in the Great Recession. More debt but interest payments are the same.
The story, however, is not so simple. Check out the timing, particularly of the rise in total interest payments. They proceed the rise in debt. Are interest payments causing debt? Actually yeah.
That parabolic increase in interest payments in the late '70s came not from rising debt-to-GDP, but from this guy.
Paul Volcker was President of the NY Fed during the mid-'70s when insiders started debating whether Arthur Burns, Fed Chairman from 1970 - '78, was corrupt or simply incompetent and when there was no debate about the competence of Bill Miller, Fed Chairman from 1978 - '79. Miller, I believe, may be the only Fed Chairman to have the majority of his own Board of Governors vote against him at an FOMC meeting.
Volcker's clout on the FOMC grew the entire time and when he became Fed Chairman in 1979, he dropped the hammer.
The Fed drove the funds rate up from 5% in 1976 to 19% in 1981. I know, I know, Scott Sumner is scowling. Here is the growth rate of NGDP over the same period.
Trending upwards, towards 15%, through the 60s and 70s. Pausing in 1976 and then in two big slams going all the way to a low of 3% in 1982.
I don't think many people doubt that the policies of the Volcker Fed drove interest rates up and inflation rates down during the early 1980s, but since the general topic of government debt is so contentious, it's worth reviewing.
However, back to our story. Paul Volcker tightened monetary policy, interest rates spiked and NGDP subsequently fell. This feeds right into our graph of Federal interest payments and debt.
Interest payments rise first as Volcker tightens. Debt-to-GDP doesn't start to rise until the Reagan double-dip. That- not-coincidentally- is exactly when NGDP was falling the farthest and fastest.
Now some people will want to point out that Reagan tax cuts as well as defense spending added to the deficit during this period and that is true. However, I want to harp on the fact that it's just not that important. Debt dynamics rules all.
It's the trend break in NGDP growth that causes the debt-to-GDP ratio to increase. Let's chart.
What we have here in blue is the log of NGDP. Why log. Because if NGDP were growing at a stable rate then the log of NGDP would be a perfectly straight line.
In red is the debt-to-GDP ratio. I marked three points where the debt-to-GDP changes underlying dynamic, in dramatically unfavorable way. All three of these points show a trend break in NGDP. And, conveniently enough, of three different natures.
In 1981, NGDP growth has a textbook trend break. It's almost as if you just put a crick in the series. Steep growth from 1966 to 1981. Shallower growth from 1981 on.
This is basically how debt crises happen. Total spending slows sharply, and revenue growth collapses - in this case taxes- but it could be sales for a business, wages for a worker, etc. However, those debt contracts don't change terms. The principle and interest agreed to were agreed to with the expectation that NGDP would continue along at its current pace. As a result, debt-to-GDP exploded.
The key here is realizing that if you just roll over debt, then the interest rate is the growth rate of the debt. Growth rates compound. If revenue growth is not in line with the interest rate, then debt and revenue will diverge at an exponentially increasing rate.
This snowball effect means that adjustments to the level of revenues or even the level of debt are not likely to have major effects once the dynamics are even a bit out of line.
I think this is what makes people feel confused and helpless in the face of mounting debt. Because, with each passing day your efforts are more and more futile. You try harder and harder but things only get worse.
To cognitively normal people, it seems like the world shouldn't be this way and this is frustrating. I can sympathize, but try as I have over the years, I cannot empathize. It all seems very ordinary to me.
The debt-to-GDP ratio responds to the 1981 NGDP in the normal way. It is trending down into 1981 (NGDP growth has actually been accelerating slightly up to this point.) Then suddenly debt-to-GDP breaks north, not stopped until the mid-1990s.
In 2001, we have what looks like is going to be a trend break in NGDP growth, but then slowly and steadily the series pulls back up to trend by about 2005. Debt-to-GDP hits a minimum in 2001, starts to climb and then begins to rollover in 2006. One can be forgiven for not having noticed the emerging rollover because of what happens next, nonetheless the slowdown in debt-to-GDP growth is smooth and clear. That's a real to top we are looking at.
In 2008, however, NGDP takes a complete downshift. Not just slower growth but negative growth. The crick in the series is practically a right angle. NGDP pulls up hard by the end of the 2009 but it is left riding well below its peak.
This is interesting from a debt dynamic standpoint because this is not the way systems usually behave and so intuition is sparse. The huge spike in debt-to-GDP is natural of course, but what next?
The slope of the log curve is roughly the same as it was in 2001-2007. That suggests that debt-to-GDP would soon top and start to decline again. However, assuming the NGDP path is permanently below its previous trajectory, the possibility for some tertiary dynamic to emerge remains.
While there isn't the traditional exploding divergence from a trend break, the integral underneath the curve is growing and I believe, exponentially. So, we do actually have a buried exponential term here. I would suspect nothing comes of it, but it does make the situation interesting.