There is an ongoing debate here in the United States about our federal debt. Obviously, we cannot keep raising the debt-to-GDP ratio, and although the federal deficit has shrunk dramatically in the past couple of years, there is a strong likelihood that we will return to growing deficits some time beyond 2018. This obviously means that the debt will accelerate again; what will happen to the debt ratio is a question for future inquiry.
As things look now, the U.S. economy is slowly rising out of the recession at growth rates 2-3 times what the Europeans are seeing. That is somewhat good news when it comes to our debt ratio, a variable that has more than symbolic meaning. Countries with high debt-to-GDP ratios pay more on their debts than countries with low ratios. The reason is simple: a country with a low debt ratio is more likely to have enough of a tax base to both fund its current spending and meet its debt obligations. GDP, obviously, is the broadest possible tax base, so the larger it is relative government debt, the safer it is to buy a country’s Treasury bonds.
The next step in this reasoning would be to ask if the debt ratio itself has any relation to GDP growth itself. In other words, does the burden of government debt on an economy slow down its growth? If the answer is yes, then rising debt creates a vicious circle including higher interest rates, the need for higher taxes and stagnant growth.
Many would say that this vicious circle obviously exists and that no further investigation into the matter is needed. However, those who say so disregard the fact that the United States, with a debt ratio above 100 percent of GDP (we cannot count just the debt “held by the public” because all debt costs money one way or the other) has a faster-growing GDP than the EU does, where the aggregate debt-to-GDP ratio for all 28 member states is 87 percent.
Therefore, as always it is good to take a look at some data. The following figure reports Eurostat data for 27 EU member states (excluding Croatia which became a member just this year) over the period 2000-2013. The data is broken down to quarterly levels and not adjusted seasonally (this vouches for “genuine” observations). The left vertical axis reports debt-to-GDP ratios while the right axis reports inflation-adjusted GDP growth numbers, quarterly over the same quarter the previous year. Since this gives us a very large number of pairs of observations, the data is organized into deciles. Each contains 148 pairs of observations – debt ratio and GDP growth for the same quarter – except for the last decile which contains 149 observations. Each decile reports average numbers for each variable for that decile:
*) The astute observer will notice that I am only reporting 1,481 observation pairs when 27 countries observed over 14 years, four times per year, should actually produce 1,512 observation pairs. The lower number reported here is due to two factors: only one data series is available for the fourth quarter of 2013, and both series for Malta are missing for the first few quarters.
While this is not an actual econometric study (that would take a lot more time than I have on my hand for this blog) the analysis nevertheless reports an interesting correlation. First, when the debt ratio rises above 60 percent, growth slows notably. The 60-percent debt level is often referred to in the public debate over government debt as a threshold governments should not cross. I have sometimes dismissed this level as arbitrarily chosen, and I maintain that any simple focus on this ratio for legislative purposes is indeed arbitrary. In fact, if we look at the other end of the spectrum a debt level below 40 percent appears to have very strong positive effects on growth. If we are going to have legislation about a debt ratio cap, then why not use 40 percent?
That said, the observed correlation calls for deeper investigation. Unlike some simplistic pundits (you know who you are…) I am not going to draw the immediate conclusion that high debt ratios cause low growth. Let us remember that GDP is the denominator of the debt ratio; if the denominator grows slowly for any reason, and government keeps deficit-spending as usual, then the debt ratio is going to rise for purely arithmetical reasons. However, as mentioned earlier, large deficits themselves can very well drag down GDP growth, raising the debt ratio for causal reasons.
More on that later.For now, let’s conclude this little exercise with two questions that I hope to answer soon:
1. Is there a correlation between large debt and big government spending? If so, the low growth in high-debt-ratio countries could have its explanation.
2. What happens if we delay one of the two variables one quarter? This classic, basic statistical method could tell us a lot about the causes and effects between debt and growth. I am going to take a stab at it as soon as time allows.
Needless to say, any future inquiry would have to include the United States. This one does not, simply because the raw data used here did not include U.S. numbers. Now that I have this data in a configured file of my own it is easy to add U.S. data.