Tyler Cowen asks us to consider the inferential value of Jamaican economic data:
It seems nothing good is pending, economic growth is negative, and the debt to gdp ratio is 143%. I take it we can agree this is one case where stimulating nominal demand will not bring much in the way of dividends? Do we agree? Last year the inflation rate was over ten percent and the (nominal) exchange rate hit new lows. Do note the country ran a primary surplus last year and is attempting to move toward a balanced budget, so does anyone wish to pin this mess on fiscal austerity? Or is their austerity and its observed failings a symptom of other policies which went badly wrong?
The measured government budget deficit is about 6.1% of gdp, but I suspect if you start the calculation in 1990, in “cyclically adjusted terms” Jamaica will appear to be running a huge surplus and a very tight fiscal policy. After all, I do see unemployment estimates in the range of 13 to 14 percent. Isn’t that a classic sign of deficient aggregate demand?
I do wonder if we can agree on this case. And if we agree on this case, which more general lessons might we draw about the difficulty of inference from data…?
There are big, econometric hurdles I’ll get to, but I think there is strong reason to suspect any demand-side (or nominal) explanations to Jamaican misery. According to this paper by Taffi Bryson at the Bank of Jamaica, unemployment seems to have fallen below 10%. But I doubt that unemployment in Jamaica is a particularly illuminating figure at all – at least with regard to demand shortfall:
The remainder of the period under study was characterised by a declining unemployment rate ranging between 13.1 and 9.6 per cent while inflation was more volatile and ranged between -0.3 and 7.9 per cent. For the sample period, the unemployment gap can be assumed to have averaged close to zero as the inflation data does not exhibit explosive behaviour (Laubach, 2001).
Fiscally, Jamaica closely approximates India. Large budget deficits accompanied by irritating inflation and bad unemployment. (Monetarily, Jamaica doesn’t borrow in its own currency, but comparisons to the Eurozone might not be justified). I think few would take either Jamaica or India to be examples of subdued demand. Indeed, cyclically-adjusted unemployment doesn’t seem nearly as bad, despite severe cyclically-adjusted austerity.
The econometric difficulties of calculating NAIRU is discussed at full-length in Bryson (2008), but needless to say relatively weak data and association should be considered with strong qualification for any policy outcome. That said, we have strong theoretical reason to believe natural unemployment in Jamaica might be high. Income GINI in Jamaica is 45.5, only slightly higher than the US, but different because:
- Jamaica should be going through a sectoral rebalancing after liberalization. Its hilly landscapes suggest agriculture is not its comparative advantage, yet relative domestic production increased in this 1990s.
- Financial sector deregulation should imply deeper credit markets as measured by total debt to GDP ratio, this has fluctuated to a lower mean.
- There is a strong expected tradeoff between unemployment and inflation, suggesting Jamaica is operating near potential.
(From Bryson, we can see the unemployment-inflation tradeoff).
Given these constraints, its possible for strong output to coincide with bad unemployment. That is, if all its income is concentrated in one man, unemployment can be startlingly high without any severe demand shortfall. This is a contrived example, but the relatively high inequality in Jamaica is just a logical extension of that argument. Indeed, even in the United States I’ve doubted whether unemployment rates mean much in the context of rebalancing towards low-paying jobs.
Jamaica also has severe supply-side issues. Its labor force participation rate, for males and females, peaked around Bill Clinton’s first election and has been in secular decline since. This explains an increasing wage rates beyond productivity. Indeed, real wage growth aspirations have beat productivity, signaling expected future inflation. In a standard AD-AS model, any idea of “demand shortfall” can’t be explained in the context of inflation to the scale Jamaica faces, without a ridiculously bad supply-side shock. Oil and labor indicators don’t indicate this. (Or is AS downward sloping?)
Which is why when Cowen asks, “can we learn anything from this data”, I think the answer is a resounding yes. The assumptions we need to tolerate for this to be a nominal phenomena, hence encouraging less austerity, far outweigh the relatively simple supply-side explanation. He warns us not to compare Jamaica with the United States, lest we become dunces, but the analogy to Italy should be clear.
Jamaica is an indebted country undergoing severe fiscal consolidation and bad growth, yet managing to run a primary surplus. Italy, too. Unlike many other Eurozone countries, Jamaica and Italy can default on their obligations without facing any immediate repercussions vis-a-vis standard of living. Default is subgame perfect, and perhaps even preferable to inflation.
The IMF released a study which contradicts intuition: that in the long-run default has little if any economic costs, and debtors aren’t restricted from capital markets. This is to say, defaulting is acredible threat, in other words: profligacy is subgame perfect.
It’s not surprising that defaulter nations aren’t long-restricted from capital markets. As Adam Smith concluded, division of labor (and hence prosperity) is limited by the extent of a market. Capitalism is an inclusive game, and it doesn’t make sense to exclude millions because of profligate government.
For countries with high borrowing rates, this strongly raises the value of formal debt restructuring. If the market believes that the Italian government thinks it can credibly-threaten to default on its debt, one of two things can happen. Rates can soar, making default inevitable, or creditors can reach a preemptive restructuring agreement offering lower rates, lines of credit, with a promise to repay debts.
The Guardian piece Cowen refers, then, is seriously out-of-touch with its anti-liberal cynicism, suggested by scare quotes around perfectly acceptable words like “owes” and “rescue”. The Washington Consensus is clearly helping Jamaica maintain credibility in international markets.
Here are a few caveats to my argument. Mainly, a paper from Irella Blackwood a (it seems) student of Brad DeLong:
This study explores the relationship between the incomes of public school teachers in St. Thomas, Jamaica and inflation rates between 2000 and 2004. It investigates how individuals adjust to potential gaps in real income and inflation as well as the consequently higher costs of living. Estimates of teachers’ incomes were gathered through a survey. The work also investigates the general awareness of inflation increases and measured the effects of high inflation on teachers. Moreover, the failure of wages to keep up with inflation over the past years has resulted in increased migration rates of teachers from Jamaican (brain drain), which will decrease the quality of Jamaican public schools, and lower the future productivity levels of Jamaican workers. This research adds to the debate over what macroeconomic policies are more appropriate for the Jamaican economy, and possibly for others third world countries. In third world countries like Jamaica, wages typically do not fully adjust to inflation. This lack of adjustment has had an extremely negative effect on the real incomes of teachers of Jamaica, and it is likely that similar negative effects extend to the entire general population.
But, as economists often say, inflation is a self-fullfiling prophecy. The consistent above-productivity expectations for wage growth are in no small part to blame. Another idea comes to mind, wherein there’s a sort of “sticky” wage growth people expect because of inflation, a self-imposed income trend if you will. I’ll play out the dynamics of this implication in another post, but for now it’s worth noting the Jamaican labor market might not be in equilibrium. But again, the long-run effects of a poor education system would support my hypothesis that Jamaica has fundamental supply-side problems.
My answer to Cowen is that while we can spin the Jamaica story in any way we want (or, as Scott Sumner insists, the Britain one too) – the strength of the assumptions in either way are what must dictate our beliefs and hence policy prescriptions. That Jamaican unemployment indicates a demand shortfall is just too improbable under this calculus.