Hacker News Reader

Comment by fauigerzigerk | Hacker News Reader

It's a little difficult to pin down the numbers right now because everything has changed so much in the last year. But 95% vs 400% is definately utter nonsense.

This IMF paper (http://www.imf.org/external/pubs/ft/spn/2009/spn0925.pdf page 35) from Nov 2009 projects US and UK government dept for 2010 to be 94% and 82% of GDP respectively.

The US and the UK, in spite of the many differences, have one important thing in common. In both cases debt is almost exclusively denominated in domestic currency. So neither of the two can default outright as long as there is enough energy left to start the printing presses. Any default would come in the shape of inflation.

View on HN

After you figure out the difference between total debt and government debt, you can look for my other comments that show that the quoted 400% figure is indeed reasonable.

The figures are apples and oranges. I have no doubt that you can get to a reasonable number of 400% for some form of debt. But if you do that you need to do something similar for all countries you compare.

What makes you think that people don't do something similar for other countries?

I don't have direct access to the necessary statistics, but based on the figures at http://en.wikipedia.org/wiki/List_of_countries_by_external_d... and guessing that most of Greece's debt is externally held, it is looks like Greece's total debt load should be somewhere in the 50-60% range. That looks a lot better than 400%. And fits perfectly with the article's contention that the Greek government has a debt problem, but the rest of Greece does not.

The list you point to includes only external debt and combines private and public debt. What that says is simply that US and Japanese citizens borrow at home whilst many others borrow abroad. It's a reflection of the size of a country's debt market and says absolutely nothing about the likelyhood of a government defaulting on its debt.

The list I pointed to is simply the first thing I came across that could be used to estimate the total indebtedness of Greece. You are right that it is not a direct measurement of total debt. You are also right that total debt is not a direct measurement of government debt risks.

However one thing that total debt is good for is seeing how much total weight the economy as a whole is carrying from accumulated debt. From the point of view of the economy it doesn't matter that much who holds the debt. After all money lost to interest is lost just as much whether it is an individual paying direct interest payments or the government raising taxes to pay its interest payments. And besides, debt can quickly change categories. For instance in a credit crisis private debt can suddenly become public debt instead. (In a recent PIMCO article Bill Gross pointed out that in credit crises over the last 300 years that public debt tends to approximately double.)

By any comparison with history or other countries, a 400% debt to GNP ratio is very high. It might not spell automatic disaster, but it suggests some rocky times ahead.

I do agree with much of what you say. What I'm taking issue with is using external debt as a proxy for the total debt load affecting an economy. Large countries with a large domestic debt market will always look better in that comparison even though the actual debt load may be exactly reversed.

I think a much more useful comparison would be to ask what share of personal and government income has to be used for debt servicing and what the trajectory of that number is.

If I had a better source to estimate total debt from for Greece I would have used it. But that was the best that turned up in a couple of minutes of Googling.

You are absolutely right that the debt servicing number is a much better thing to use. It not only handles obvious things like different interest rates for different entities, but also subtle ones like preventing double-counting of securitized debt. (In securitization a pool of debt is sold to a company created for the purpose that immediately issues bonds which sum to the total debt modulo trivial operating expenses. Thus securitized debt shows up as both personal and corporate debt.) But that information is harder for a random internet user like me to get at than total debt numbers.

But better thing to use or not, it is harder for me to estimate the total cost of servicing the debt. I'm just a random person on the internet and don't have direct access to any of the data. And besides, total debt is what the original article used, so an apples to apples comparison talking about what was in the article should use total debt.

"total debt is what the original article used, so an apples to apples comparison talking about what was in the article should use total debt."

I agree that total debt or some proxy for it should be used for the comparison, but the numbers on the wikipedia page you linked as well as (apparently) the numbers in the article are _not_ total debt. Not even close. These numbers show only external debt.

That excludes almost all US mortages, it excludes almost all US and Japanese government debt, whereas for other countries it includes almost all debt that anyone, private or public, has taken on. That's why comparing these numbers says absolutely nothing about the total debt load an economy has to support.

External debt does matter in some respects, it just says absolutely nothing about total debt, which is what you say should be compared. It's not just a little bit off. For instance, Japan's government is indebted to the tune of more than 200% of GDP. Yet, external debt of Japan as a whole, including private and public debt, is just 35% of GDP according to that wikipedia page.

as long as there is enough energy left to start the printing presses

Intuitively this makes sense but in reality it is not practical. As soon as investors hear the presses crank up (metaphorically speaking) they can predict what's coming and will dump all their US debt, driving up interest rates and forcing a deep[-er] US recession. Politics will then correct this before the USA becomes Northern Zimbabwe.

I'm not worried about the US becoming Zimbabwe at all (northern or otherwise). But under extreme circumstances, when the alternatives are to default outright or to inflate the debt away, the latter will always be preferrable for a country indebted in its own currency. It's not an option for other countries, so I think it's a useful distinction in spite of the issues you correctly point out.

Reply on news.ycombinator.com