June 09, 2013
Brad DeLong has a set of ruminations on the economic situation that we now face, the gist of which is that we better be cautious in using fiscal policy because “we” are worried that the bankers may sink themselves again if interest rates rise back to more normal levels. Let’s look at the argument starting with figuring out who “we” is.
While Brad never tells exactly, when he says “we” he is obviously referring to the mainstream of the economics profession. And his “we” actually got considerably more wrong than he gives them credit for in his post. To start with, “we” hugely overestimated the effectiveness of monetary policy in sustaining full employment in the United States. In citing the Fed’s successes Brad tells us:
“There was the 2001 collapse of the dot-com Bubble that took $5 trillion of equity wealth down with it. In all of these cases the Federal Reserve was able to react swiftly and smoothly to keep these large financial shocks from having much of an effect on the real economy of production and employment.”
That’s not what the data show. First of all, from peak to trough we lost $10 trillion in equity value. The crash went far beyond the dot-coms, the stock price of everything from GM to McDonalds plummeted in the stock market crash of 2000-2002.
It turns reality on its head to say there was not much effect on employment from this crash. Employment peaked in February of 2001. It didn’t cross this peak again until February of 2005. The employment peak in the private sector was reached in January of 2001. It didn’t cross this again until June of 2005. In both cases this was at the time by far the longest stretch without gains in employment since the great depression.
Looking at employment to population ratios, the EPOP peaked at 64.7 percent in April of 2000. Five years later it was a full 2 percentage points lower. That corresponds to 4.4 million fewer people holding jobs.
It is also worth noting that the federal funds rate was lowered to 1.0 percent in the summer of 2002 and remained at this level for almost two full years. This was the same rate that the ECB had in place following the 2008 meltdown until it recently lowered its overnight rate somewhat further. The 1.0 percent ECB rate was already thought to be pretty much a zero lower bound in the sense that no one believes that further reductions have any substantial payoff in economic activity. For practical purposes the Fed had hit the zero lower bound following the 2001 downturn.
In other words, it was easy to see from looking at the data that recovery from the stock market crash induced recession in 2001 was not easy. It is scary to think that “we” did not recognize that fact at the time and apparently still do not recognize this fact today.
Since the recovery from the 2001 stock crash was difficult, it should have been easy to predict that recovery from the collapse of the housing bubble was going to be difficult. The housing bubble was driving the economy ever since the stock crash.
Bubble driven house prices had raised construction to near record levels. The share of GDP going to residential construction was more than 2 percentage points above its historic average. It was 100 percent predictable that we were going to lose this boom when prices collapsed. Also since the boom led to overbuilding, which manifested itself in the form of record high vacancy rates, it was certainly predictable that construction would fall back to below normal levels. As it was, we lost more than 4 percentage points of GDP in the decline in residential construction following the crash ($640 billion in today’s economy).
The equity created by the bubble also led to a wealth effect driven consumption boom with the saving rate falling to near zero. (I would argue that capital gains showing up as income led to an overstatement of income in this period, so the actual saving rate was even lower than the reported saving rate.) In any case, it was fully predictable that consumption would plummet following the loss of $8 trillion in housing wealth associated with the collapse of the housing bubble. If we assume a wealth effect of 5 cents on the dollar, that translated into a loss of $400 billion in annual consumption. (That was also a bubble in non-residential real estate, but we’ll ignore this for the moment.)
This means that the combined loss in construction and consumption spending led to a drop in annual demand of more than $1 trillion. What exactly did the Fed have in its bag of tricks that was supposed to replace this amount of demand? What sector(s) of the economy was so sensitive to interest rates that it could plausibly expand to fill this huge gap in demand? Certainly there was nothing that was in evidence as of 2007. Hence “we” should have been able to see that the collapse of the housing bubble posed very serious risks to the economy.
Next we have Brad giving us warnings that we still need to take Reinhart & Rogoff’s debt growth trade-off seriously. It’s not clear what empirical work he is relying on in pushing this contention, but it certainly is weak at best (e.g. see here and here). As I have also argued as a theoretical point the idea that there can be any substantial growth penalty from debt per se seems absurd on its face.
Even though our debt to GDP ratio is the highest it has been since the early post-war years, at around 1.0 percent of GDP the interest burden is at a post-war low. We aren’t looking at getting back to the early 1990s levels of more than 3.0 percent of GDP for more than a decade. And, even that burden did not preclude a decade of solid GDP growth in the 1990s. If there is in fact a debt-curse then we can buy back debt at steep discounts when interest rates rise, as is predicted by the Congressional Budget Office and other forecasters. We can also sell off assets to get a super-normal return in the form of an extra growth dividend.
But all of this is a side-bar to the highlight of Brad’s story:
“If you are Marty Feldstein, you think there is a 20% chance of normalization of interest rates in the United States in each year looking forward, that because of large debt outstanding, the normalization of 10-year Treasury rates carries them not to 4% per year nominal but to 6%, which means that should normalization come that’s a 36% capital loss on bank and shadow bank holdings of 10-year Treasuries and other things of equivalent duration. Given underlying political currents it is not at all clear that the political support for a second rescue of the banking system anywhere exists in any North Atlantic democracy. Hence Stein, Feldstein, and George draw the conclusion that it is time to start raising interest rates so banks can get a wedge between the zero cost of funds from depositors and the amount of money they earned by holding safe medium term assets. Otherwise, they fear, we will soon face a financial crisis worse than 2008-9.”
Okay, in case this is not clear, Brad is attributing a view to Martin Feldstein and others that the banks are insufficiently hedged against the risk of a rapid rise in interest rates. Therefore, we should start raising interest rates now (with a hit to growth and jobs) in order to avoid a sudden rise in rates that will send the banks back into insolvency.
Let me put this another way just to make the point more clearly. Martin Feldstein and his ilk supposedly believe that the banks have again put themselves at great risk. If the government (Fed) doesn’t take actions to slow the economy today, then the banks may again be put into bankruptcy.
Now Brad sees this as a horrible story for all of us because he seems to think that if the banks go under again there will be nothing that we can do to get the financial system back on its feet. I would argue otherwise, since central banks like the Fed have almost limitless power to re-inflate the economy. (Yes, there are political constraints, but I trust that if faced with the prospect of enduring double-digit unemployment we will be able to overcome the political constraints. Note that contrary to what is often claimed, this is not a one-time deal. If the banks collapse we can restart them a week later, a month later, or even five years later. It is best that the banking system not collapse, but we do know how to re-inflate the financial system and the economy.)
But let’s just focus on the key claim, Brad thinks it is plausible that the banks have again made themselves vulnerable so that a plausible (in his view) set of economic events could wipe them out. Is this to be taken seriously? After all, President Obama assured us that Jamie Dimon and Lloyd Blankfein are very savvy businesspeople (the CEOs of J.P. Morgan and Goldman Sachs, respectively).
As a practical matter I have no idea how vulnerable the major banks are to a rapid rise in interest rates (which I view as highly unlikely), but if they are really that vulnerable, how do we justify letting their CEOs get paid $20 million a year when, if Brad is taken literally, they have put the whole economy in jeopardy yet again. In other words, if we take Brad at his word, we have to keep millions of people out of work because Jamie Dimon and Lloyd Blankfein have made their banks vulnerable to actions that would boost the economy, and if their banks go down, the economy goes down.
I think Brad is 100 percent wrong about the economy’s need to keep these banks afloat and that Feldstein is almost certainly wrong about the risk of a sharp rise in interest rates, but if we assume both are right, then it is the best argument yet for breaking up the big banks and radically altering the structure of banking regulation.
Note: Thanks to Robert Salzberg for correcting typos.
Comments