Winner of the New Statesman SPERI Prize in Political Economy 2016


Sunday 10 August 2014

Is pessimism about European debt levels justified?

Barry Eichengreen and Ugo Panizza posted a rather pessimistic account of the sustainability of European debt levels. To quote:

“For the debts of Europe’s problem countries to be sustainable ... their governments will have to run large primary budget surpluses, in many cases in excess of 5% of GDP, for periods as long as ten years. History suggests that such behaviour, while not entirely unknown, is exceptional.”

Is this pessimism warranted? Here are some numbers, updating an earlier post.



Net debt % GDP
2013
Interest % GDP
2013
Implicit nominal rate
2013
Long term
 r-g (high)
Required primary surplus % (high)
Long term 
r-g
(low)
Required primary surplus % (low)
Underlying Primary surplus
96-07

Underlying
Primary
surplus
2015

Greece
122.7
3.6
2.9
5
6.1
2.5
3.1
0.0
7.8
Portugal
91.8
3.8
4.1
5
4.6
2.5
2.3
-2.0
4.7
Ireland
90.3
4.0
4.4
5
4.5
2.5
2.3
1.1
3.0
Spain
70.7
2.9
4.1
4
2.8
2
1.4
0.9
0.2
Italy
116.5
4.9
4.2
4
4.7
2
2.3
2.4
4.9
France
73.6
2.1
2.9
2
1.5
1
0.7
-0.8
0.9
Germany
49.1
1.6
3.3
2
1.0
1
0.5
0.6
0.7










Japan
137.5
0.9
0.7
2
2.8
1
1.4
-4.4
-5.4
UK
65.4
2.8
4.3
2
1.3
1
0.7
-0.1
-1.4
US
81.2
2.3
2.8
2
1.6
1
0.8
-0.2
-1.7










Euro area
68.5
2.5
3.6
3
2.1
1.5
1.0
0.8
1.8
OECD
69.1
1.9
2.7
3
2.1
1.5
1.0
-0.1
-0.8

All data comes from the OECD’s Economic Outlook. The first three columns are self explanatory. The fourth column is a complete guess at what a long term growth corrected real interest rate (r-g) might be, and I will discuss these numbers below. If you multiply this by the debt stock, you can compute what the required primary surplus needs to be just to keep the debt to GDP ratio stable. To start getting debt down, surpluses would have to be larger still.

Therein lies Eichengreen and Panizza’s pessimism. They argue that long periods over which primary surpluses have been above 3% are rare. Of the PIIGS, only three of the five are expected to have a primary surplus above the required level by 2015. Are primary surpluses above this required level possible to maintain for a decade or two rather than a year or two?

Yet a quick look at the table shows you that the problem is not so much the starting level of debt, but the assumption about trend r-g. Halve r-g, and you halve the required primary surplus, as columns 6 and 7 show. Numbers above this level, although still a challenge, look much more possible.

The interest rate numbers in the high column assume 2% is the risk free value for r-g, corresponding to a 4% real interest rate and 2% trend real growth. It is applied to the US, UK, Germany, France and Japan because there the chances of default are minimal. We then add percentage points for risk. The argument here would be that, within the Eurozone, OMT prevents numbers getting very large, but equally we will not return to the pre-2008 days when Greek debt was considered only marginally more risky than German debt.

However, I think you could quite plausibly divide all these numbers by two. A number like 4% for the risk free real interest rate would apply to some earlier decades, but at the moment this number looks rather high. [1] Secular stagnation could keep this number even lower. Risk premiums for the high debt Eurozone countries could also be halved, as they are anyone’s guess given institutional uncertainty. With these low numbers, required primary surpluses become more feasible.

The importance of the interest rate assumption also becomes clear if we compare the Euro area to the OECD as a whole. Looking at debt levels and actual primary surpluses, the one country that really looks worrying is not in the Eurozone, but Japan. In terms of primary surpluses, neither the UK nor US is any nearer achieving required levels than the PIIGS. The only reason to single the PIIGS out is because of the interest rates they might have to pay. We saw with OMT that this has as much to do with the institutions of the Eurozone as a whole as it does individual governments. 



[1] The CBO estimates (pdf, page 92) that the real interest rate on 10-year US Treasury notes averaged about 3 percent during the 1960s, about 1 percent during the 1970s, about 5 percent during the 1980s, about 4 percent during the 1990s, about 2 percent between 2000 and 2007, and about 1 percent during the past six years. CBO projects that the average real interest rate the federal government will have to pay on all its debt from 2014 to 2039 will be 1.7%, corresponding to a real rate of 2.5% on 10 year bonds (p104). This implies numbers for r-g perhaps even lower than the ‘low’ column in the table.


5 comments:

  1. "Looking at debt levels and actual primary surpluses, the one country that really looks worrying is not in the Eurozone, but Japan."

    We should not look at it this way. Japan is a net-creditor in the international system. It has a large private sector surplus. If it chooses to have its surplus savings channelled into large government debt, which is invested in things like education and health that is its business. It does not owe money to others in other currencies. Even if its private surplus runs down as it ages, so what? That is what it has been stocking up for.

    The crisis hit Eurozone periphery is another matter. But the debt itself is not the problem. It is a symptom. They have a terms of trade problem which means they have an external payments problem.

    The only solution for these countries is fiscal union and large scale investment with German capital because they do not want a return to the bad old days of devaluations (liras and drachmas).

    ReplyDelete
  2. Important post in the line of a recent book on Portuguese government ability to pay the public debt service http://www.bnomics.pt/products/vamos-conseguir-pagar-a-nossa-divida-publica. However, I would prefer to use Eurostat data concerning gross public debt http://epp.eurostat.ec.europa.eu/tgm/table.do?tab=table&init=1&language=en&pcode=tsdde410&plugin=1. Congratulations

    ReplyDelete
  3. For advocates of Modern Monetary Theory, this is all a storm in a tea cup, and for the following reasons.

    1. If the private sector wants to hold a larger stock of government liabilities than in recent decades, then government should let it, and pay a miserable rate of interest. Indeed, if government DOES NOT supply the private sector with those private sector assets / government liabilities, the private sector will go into Keynsian paradox of thrift savings mode, which will raise unemployment.

    2. Alternatively, if the private sector DOES NOT WANT to keep an elevated stock of government liabilities, it will try to spend it away, which will raise demand excessively. In that circumstance, government can run a surplus without causing a recession. In fact government will be more or less FORCED to run a surplus in order to contain inflation.

    Problem solved. In fact in a sense, governments ARE FORCED to adopt the above MMT policy and just ignore the whining that we get from the Eichengreens of this world. Incidentally, the above points are relevant for monetarily sovereign countries and the EZ as a whole. INDIVIDUAL countries within the EZ are slightly different.


    ReplyDelete
  4. Also worth thinking about why difficult to get primary surplus up. Depends on standard thinking about tax levels and sources. See http://www.futureeconomics.org/2012/01/austerity-is-political/ and http://www.futureeconomics.org/2012/07/keynes-is-all-we-need/

    ReplyDelete
  5. One thing to note here is that we are talking about debt and the calculations done here assume implicitly the a primary surplus is either constant or will be distributed in its entirety no matter how high it is which is obv not the case. so the expected primary surpluses should be a little higher to ensure stability of sustainability.

    ReplyDelete

Unfortunately because of spam with embedded links (which then flag up warnings about the whole site on some browsers), I have to personally moderate all comments. As a result, your comment may not appear for some time. In addition, I cannot publish comments with links to websites because it takes too much time to check whether these sites are legitimate.