Debt to GDP ratios are irrelevant (or : why Japan is not in such a difficult fiscal position)

Over the past few years the “Japan is going to default” theme has proven quite popular both in blogs and regular financial media. It’s true that the headline number (over 200% debt to gdp !) is eye-catching – as it’s 3 time more than most developed nations.

But this number is also very misleading, and this post will (hopefully) explain why Debt to GDP ratios are irrelevant when assessing a nation’s likelihood to default. I’ll use two groups of countries to illustrate the points made : Japan and Greece (very high debt to gdp), and Germany, Canada & USA (similar and average debt to gdp), to show that they all have very different risk profiles.

Sovereign or not ?

The first point to check is not a number, but a qualitative point : is the country’s debt issued in a money it can freely print ? If yes, a default can only happen as a political decision (ex : Russia 1998), as you can never run out of a commodity (domestic money) of which you are the monopoly supplier. Obviously debt monetization is not friendly to investors, but is not considered a default, so this distinction is important (think CDS contracts).

This simple step already identifies the main problem with Greece : it can’t print euros, therefore is revenue constrained. The Greek government needs to earn the euros needed to service its debt

Note : Officially, Germany can’t print euros, and they have a very strong public stance against debt monetization. However, they do control the ECB, and could certainly print money if necessary. I believe that the market understands this, and it is reflected in Germany’s CDS spreads.

Estimating a country’s repayment ability – today

What matters to assess default risk is a country’s ability to service its debt, and debt to GDP does’nt give much information on a country’s repayment ability. Indeed, GDP is not the resource of the government – it’s the country as a whole, and neither is the gross debt the cash that the government has to pay.

Net debt (Gross debt – financial assets) to GDP is already an improvement over Debt/GDP. It’s a finer measure, helping to estimate the real debt situation of a country. For instance, Greece has no financial assets, while Japan has a lot – its Net debt to GDP is still high (117%), but nowhere as high as the headline number.

However, it still doesn’t inform about the payment ability of the country, as the cost of the debt to be serviced could be massively different from country to country. The Net Interest Service as a % of GDP (does NOT include principal repayments) shows how much of the country’s resources the debt service is using. Dividing this number with the country’s tax burden gives a reasonnable approximation of the strain of interest payments on the government. I’ll call debt burden the Net Interest Service as a % of government resources.

The debt burden highlights the bleak situation in Greece : before paying a single state employee, even before repaying debt principals, it has to pay 15% of its income, or three times more than Japan or USA.

The debt burdens of Japan, Germany and USA are roughly equivalent (respectively 4.3%, 5.4% and 6.3%), the main difference being their different tax burdens. USA and Japan have quite a lot of leeway to increase taxes, while Germany has less potential, being at 41% already (more than 50% is generally thought as counterproductive). Japan actually has the smallest debt burden of those three countries – so much for the “Japan is going to collapse under its debt” theory. Of course they have some specific problems that could harm them in the future (mainly the population aging that reduce the savings rate, therefore increasing bond yields), but as of today, Japan has a very safe fiscal situation.

Looking Forward

While the analysis above gives a solid estimate of a country’s current ability to service its debt, it is insufficient as it doesn’t peek into the future. The simplest way to “look forward” is to check the government’s structural budget balance. It shows why Germany can enjoy such low CDS spreads and bond yields : its structural budget deficit is below long-term inflation – perfectly sustainable – while Japan and USA are far above. USA is anyway an exception, as they issue the world’s reserve currency – allowing them to enjoy significant fiscal flexibility.

Who’s the next “domino” on the line ?

The table below (sorted by CDS) highlights how these two indicators allow to understand quite precisely the soundness of a country’s fiscal position. Greece has by far the worst stat, and that shows in its CDS (world’s highest). When looking at the debt burden only, Spain doesn’t look so bad , but taking in account the structural balance allows to understand why the market is not confident :

Italy is an even more interesting case : its debt burden is very high, it has no fiscal leeway, the only factor that keeps it ‘alive’ is its very reasonable structural balance. If growth slows down, Italy will become the new focal point of the EMU debt crisis – it is truly the “Elephant in the room

This is a very simplified way to take a look at a country’s fiscal situation, however it uses only public data, and allows to go a bit further than “200% debt to GDP is the end” type of analysis, in a couple of minutes. Other indicators can help detail the analysis such as estimated GDP growth, current account deficit (if negative, you need foreigners to lend you money), and debt maturity (to pinpoint the time of potential default), savings rate.

Finally, I certainly hope that you noticed that debt / gdp is a poor predictor of the actual debt burden of a country – Japan has a very safe position, and Canada – who appears to be in the same range than USA & Germany – has almost no debt burden.

All data from IMF except Net interest service as % of GDP (OECD) and CDS prices (Bloomberg)

4 Responses to Debt to GDP ratios are irrelevant (or : why Japan is not in such a difficult fiscal position)

  1. xinrong says:

    One question, what’s the % of tax income to the government’s overall income?

  2. xinrong says:

    my question is have you considered the debt income of government which could partially offset the debt payment?

    • Nicolas Paris says:

      Yes, i use the Net Interest Service data from the OECD : net interest service = cost of interest on the debt – revenue from assets

      so for Japan, the net interest service already includes the income from their US Bonds holdings

      Regarding the tax question, as you can see here, most of the government revenues are actually from taxes, so i think that the calculation used (NIS as a % of government revenues = NIS as % of GDP / tax burden) is a reasonable estimate, at least for developed (and most developing) nations.
      It wouldn’t work for natural resources rich countries though. Anyway, i don’t think any of those are in OECD (except maybe Chile), so i wouldn’t know where to find the net interest service.

  3. A reader was wondering about the exact definition of the tax burden.

    From what I’ve gathered, it’s not a theoretical number, but the actual taxes paid (individual, corporate, sales) – divided by the GDP

    Sources :

    [search for tax burden]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: