Saturday, September 21, 2013

Austerity in recessions

consolidation_growth.pngMenzie Chinn has a post on austerity in the latest recession.  Here is the graph:

This shows a correlation between economies that increased their structural government budget deficits and economies that recovered from the recession.  In other words, increased spending or tax cuts help economies recover.  He included a link to his data source, so I played around with it to see what I could find.

Here is my first graph.  I used OECD countries, which I think differs slightly from his grouping, and my scale on the y-axis is average annual real GDP growth, where I think his is the total over 4 years.  But, it looks to me like we basically have the same data and the same general correlation:

But, in the end, I find that the causal factors are probably the initial expenditure levels and structural balance.

How does the starting structural balance affect the results?

The first thing I wanted to check was how the beginning structural balance of a given nation affected growth in the recession.  Here is a correlation between the starting structural balance and subsequent growth:

This correlation is similar to the correlation for consolidation.  In other words, the higher the surplus (or lower the deficit) going into the recession, the better the economy did coming out of the recession.

First, I want to make sure that we are not just measuring the health of the economies going into the recession.  Maybe the high structural balances are still carrying some signal from recent economic performance.  So I'd just be showing that a strong economy leads to a strong economy.  So, I compared the 2008 structural balance to 2004-2008 GDP growth.  There is no correlation.  The structural balance measure really doesn't carry any residual bias related to the strength of the economy.

Additionally, in the analysis that follows, the relationships are similar regardless of whether we use real growth from 2008 to 2012 or the difference in growth between 2008-2012 and 2004-2008.

Maybe consolidation and poor growth are both caused by the beginning structural balance.

Here is the correlation between the beginning structural balance and the subsequent consolidation:

This is a much stronger relationship than the previous correlations.  This suggests that it is the beginning structural balance itself that leads to a need for consolidation.

If we look at these relationships from the 2004-2008 period, the pattern is the same, but the correlations are very weak - correlation coefficients in the low single digits and slopes of -.12 and +.10.  And the correlation between a negative structural balance and consolidation in the earlier period was much weaker.  In good times, countries that have negative structural balances weren't forced to consolidate, and consolidation wasn't as painful when they did.

So, during healthy economic times, a nation can get by with a range of budget surpluses or deficits without a significant effect on the economy.  But, this seems to clearly show that the most important element when recessionary conditions arrive is to be sitting on a positive structural balance when it happens.

Is the negative correlation between consolidation and growth meaningful?

To try to answer this question, I divided the 33 nations in my sample into 3 categories, based on their structural balances in 2008.  The low group started with a structural balance of -14.2% to -4.4% of GDP.  The middle group started with -3.8% to -1.3%.  The high group started with -1.1% to +2.4%.

Greece in the outlier here.  If I remove Greece, the trendline on the low balance group has a slope of -.39 and an r squared of .187.

The effect of consolidation on growth is negative for the countries starting with a poor structural balance, it has little correlation with the average countries, and consolidation correlates positively with growth on the countries that started with above average balances.  The effect of consolidation on growth is a product of the beginning structural balance.

The take away - Keep your structural balance in order when times are good.  If you do, then when bad times come, growth out of the recession will be generally unrelated to consolidation.  You'll have decent outcomes with or without "austerity" policies.  If you don't, then you may be forced to institute austerity when your economy is least able to handle the dislocations.

How do Revenues and Expenditures affect the importance of the beginning structural balance?

The level of government expenditures and revenues in 2008 is negatively correlated with growth from 2008 to 2012.  But the correlation is much stronger for expenditures.

In other words, the benefits of a high structural balance come mostly from lower expenditures.

In general, these relationships seem to suggest that smaller governments create more resiliency in economic downturns.  They also suggest that while fiscal stimulus might create a small boost in a healthy economy, the negative effect of the higher structural deficit probably negates it.

Countries with a large structural deficit when an economic downturn hits are probably forced into consolidation by budget constraints.  For countries without a large structural deficit, I don't see a relationship between fiscal stimulus and recovery.

Here is a regression that breaks out the benefits of a high structural balance during the economic downturn.  For each 10% reduction in 2008 expenditures/GDP, average growth from 2008-2012 was 2.1% higher.  For each 10% increase in 2008 revenue/GDP, average growth from 2008-2012 was .7% higher, although the revenue coefficient is not statistically significant.

These numbers suggest that for the recent downturn, lower initial expenditures were related to a more robust recovery while initial revenue levels were important inasmuch as they helped avoid a dangerously low structural balance.

Let's have our austerity during the recovery!

1 comment:

  1. Great post, thanks for getting into the details here.