Thursday, September 20, 2012

Two views of debt and stagnation

Two new papers on economic stagnation in periods of high government debt (i.e. now) are making a splash: 

Public Debt Overhangs by Carmen  Reinhart,Vincent Reinhart and Ken Rogoff
The Output Effect of Fiscal Consolidations by Alberto Alesina, Carlo Favero and Francesco Giavazzi

This review is mostly about the former, with a little mention of the latter (maybe I'll get back to that later)

The Reinharts and Rogoff look at episodes in which government debt crossed 90% of GDP. They have two big conclusions: the episodes lasted  a long time, "...among the 26 episodes we identify, 20 lasted more than a decade," and those episodes are associated with slow growth: "the vast majority of high debt episodes—23 of the 26— coincide with substantially slower growth."

They want very much to conclude that high debt causes the slow growth, referring to "growth-reducing effects of high public debt." But as always in economics, correlation is not causation, which they recognize:
But obvious concerns arise here about cause and effect. Is the public debt overhang causing the slower growth? Or is an exogenous shock that causes slower growth either helping to generate the public debt overhang or else prolonging the escape from that debt overhang?
Evidence? Well, the debt episodes last a long time
The long length of typical public debt overhang episodes suggests that even if such episodes are originally caused by a traumatic event such as a war or financial crisis, they can take on a self-propelling character...
 The long duration belies the view that the correlation [high debt with low growth] is caused mainly by debt buildups during business cycle recessions. ...
No, alas. This makes a pretty good first-year exam question: write down a model in which income is completely exogenous (unrelated to debt levels) yet once a country crosses 90% debt/GDP it takes decades to repay, and growth is slower conditional on high debt. (Hint: Use the permanent income model. Countries get in debt when they have bad income shocks. Debt has a unit root in that model, so debt excursions are never expected to revert.  It does take "growth fluctuations" that are beyond "cyclical," but those do exist, even without high debt.)

Ok, well,
This endogeneity conundrum has not been fully resolved. However, a number of recent studies have tackled the problem. .... [they] have concluded that the relationship cannot be entirely from low growth to high debt, and that very high debt likely does weigh on growth.
Oh, great. "Studies." Yet, as I read the review of the "studies," they are the usual sort of growth regressions or instruments, hardly decisive of causality.

I shouldn't be too hard, because I agree with the conclusion (high debt is likely to cause low growth). I'm just picky about the logic. But for a reason.

What's missing? A mechanism. To discuss cause and effect sensibly we have think about the plausible mechanism is. Regressions can too easily conclude that since rich guys drive BMWs, all you need to do is drive a BMW and you'll get rich.

And clearly, debt by itself doesn't matter -- it's how debt leads to other economic events that matters.

This is to me a frustrating feature of Reinhart and Rogoff's earlier work. Recessions after financial crises are typically longer (usually misquoted as "always.") Ok, but why? Because governments follow policies after financial crises that screw up economies for a long time (distorting taxes, wealth transfers, propping up zombie financial institutions)? Because of "private debt overhang" that would be cured by a massive transfer from savers to borrowers? (Not my favorite theory, but popular around the lunchroom so I'll mention it.)  Because the destruction of property rights in bailouts freezes new investment?  Their work is quoted as a mysterious fact of nature about which nothing can be done.

Here, Reinharts and Rogoff do mention some mechanisms
The first channel operates through a quantity effect on private sector investment and savings. When public debt is very high, it will tend to soak up the available investment funds and thus to crowd out private investment. If the government at the same time is imposing policies that attempt to reduce its debt burden with higher taxes, a burst of unexpected inflation, or various types of financial repression, then investment may well be discouraged further.
The first mechanism seems to me to confuse debt with deficits. The second one rings true: high debts correspond to high taxes (really high tax rates), wealth expropriation, and other big drags on investment. Financial repression is an under-reported issue:
In addition, governments in the second half of the twentieth century often used policies of “financial repression” to reduce the cost of the public debt, by limiting capital flows and regulating financial institutions in such a way that alternative investments were blocked and financing for government debt would flow more cheaply.
See Banks, comma, European. And given the detailed control that Dodd-Frank gives to US regulators, I can see "gee, we didn't see you at the Treasury Auction. Should we send some inspectors down to look at the books?" coming to a bank near you soon.
The second channel involves a rising risk premium on the interest rates for government debt. Sufficiently high levels of public debt call into question whether the debt will be repaid in full, and can thus lead to a higher risk premia and its associated higher long-term real interest rates, which in turn has negative implications for investment as well as for consumption of durables and other interest-sensitive sectors, such as housing. 
This makes less sense by itself. Why should a risk premium on government debt matter to private investment?  Well, because we can all see that an indebted government is going to tax away private businesses... but we already talked about that.

A mechanism could let us sort out cause and effect. We can see distorting taxation, financial repression, property rights destruction in defaults, inflation, and see which paths following high debt make growth better or worse.  (Many PhD theses here!)

And, more importantly, the correlation is really pretty useless until we figure out which mechanism is at work.

RRR's Conclusions:
This paper should not be interpreted as a manifesto for rapid public debt deleveraging
exclusively via fiscal austerity in an environment of high unemployment.
OK, but I find this annoyingly misleading. Why sign on to the deliberately obfuscation induced by current political use of the word "austerity"? Cutting spending is a lot different from raising marginal tax rates. "Unemployment" sounds like an endorsement of short-term Keynesian stimulus, which must be the one thing that clearly doesn't work in their data once debt gets to 90% of GDP.

Alesina and company make this clear:
Adjustments based upon spending cuts are much less costly in terms of output losses than tax-based ones. Spending-based adjustments have been associated with mild and short-lived recessions, in many cases with no recession at all. Tax-based adjustments have been associated with prolonged and deep recessions. 
Here we have in a nutshell my frustration with the Reinhart-Rogoff paper. There is a causal mechanism staring us in the face -- high taxes, prospective wealth confiscation (and financial repression) kill growth. Yet, they want to make "debt" the culprit, not really looking at the causal mechanisms in any detail. Why are they not just a big data set for Alesina and co's conclusions?  Back to RRR:
Our review of historical experience also highlights that, apart from outcomes of full or selective default on public debt, there are other strategies to address public debt overhang including debt restructuring and a plethora of debt conversions (voluntary and otherwise). 
Now you get the agenda and weak discussion of causal mechanisms. If "debt" is the problem, the answer is obvious: default or inflate it away. "Restructuring" and "conversions" are nice words for default.

But the case for default is not, in fact, made anywhere in the "review of historical experience" in this paper. Serial defaulters in their data do not have higher growth rates. Paying it back worked out OK for Alexander Hamilton. The Soviet Union was inaugurated the opposite way with a big default. If washing your hands of debts is such a good idea, it's interesting that so many governments go to such lengths to avoid it.

Where is the option, liberalize your economy, and grow out of it? They dismiss the one great data point that goes against the trend, the UK paying off Napoleonic war debt, thus,
there were substantial transfers from the colonies to finance debts and facilitate debt reduction...With the exception of the United Kingdom at the height of its colonial powers in the nineteenth century,

So forget  free markets, industrial revolution, railroads and all that -- England just taxed colonies like ancient Rome?

Speaking of the 19th century
In those days before fiat currency, inflation was not as prevalent as it would later become. Thus, the “liquidation” of government debt via a steady stream of negative real interest rates was not as easily accomplished in the days of the gold standard and relatively free international capital mobility as in the decades after World War II.
This sounds like a bad thing!

Yeah, default sounds great ex-post. But it is the precommitment against default ex-post that lets you borrow ex-ante. To say nothing of the chaos a large-scale sovereign default or inflation in the US and Europe would cause. Not so easy.

I don't mean to sound one-sided on this. I've been advocating Greek default for a while, at least while the original bond holders still held some of the debt. (Too late now). I'd still rather see us all  liberalize, grow, and pay it off. I'd rather see governments cut spending, as I see that paying it off by confiscatory wealth taxes will lead to a big no growth data point. Default is only a little better than that option. But let's face up to the costs of default, not just how nice it will be to wipe out the debt.
However, the evidence, as we read it, casts doubt on the view that soaring government debt does not matter when markets (and official players, notably central banks) seem willing to absorb it at low interest rates—as is the case for now.

I'm glad to end on a note of total agreement. "As is the case for now" only applies to some countries -- ask a Greek friend!