Wednesday, 12 October 2011


Note 1: I am incredibly grateful to Aristos Doxiadis for citing this article in his excellent book, Το Αόρατο Ρήγμα.

Note 2: This blog was updated on 5 Dec 2014 to reflect new data on the size of bribes

Veteran readers will remember my epic slugfest with our new wave of defaultniks, a propos of the release of #Debtocracy. A central bone of contention was the defaultniks’ claim that much of the Greek debt was not attributable to the will of the people and was in fact odious. The defaultniks purposefully refused to offer even an approximation of what percentage of the debt they considered to be ‘odious’ in this way but pointed to excesses in public procurement and public investment costs as indirect evidence. When pressed on the matter of how much of the debt is odious, they flitted from ‘all of it’ to ‘some of it, surely we deserve to know how much!’ depending on their audience in any given moment. 

I argued, on the other hand, that with nearly two thirds of all spending going directly to the people in the form of direct transfers, pensions and public sector wages, it is very unlikely that most of the public sector’s debt in Greece was odious. Still, I acknowledged that someof it probably is.

The months have rolled past and the defaultniks are by now so convinced of their moral and intellectual superiority (or at least the physical muscle they can command) that they see no point in following up on this argument. If they've managed to put together a self-styled Debt Audit Committee, answerable to no one and selected by buddy-up, it has made no attempt at a figure and will likely not attempt one until after La Revolucion. Yawn.

Government, of course, has no interest in such calculations so I can’t count on them. 

So screw everyone. I have to do this myself. Like the defaultniks themselves, I will start with procurement because that's where the bodies are chiefly buried. 

First, I need an estimate of the actual procurement spending of the Greek government, going back as far as possible. Eurostat provides this (if you bother to divide % of GDP by % of total contracts) from 1995 to 2009, and you can find a link to this and other interesting datasets here.

The result – about 9 to 13% of GDP (about a fifth to a quarter of all government spending) went on public procurement annually. On a typical year, roughly 60% of this was under the radar spending that was never published in the official procurement journal of the EU because the contracts were (whether really or artificially) too small. The estimated amounts spent on everything, from the Rion-Antirrhion bridge to felt tip pens, are as follows:

Now we need an estimate of the percentage of this that went on bribes. The World Bank generally calculates that 3.7% of all procurement spending globally is spent on bribes, but I prefer to use the percentage admitted to by Siemens, whose executives have had their own run-ins with the greasy outstretched palm of the Greek government official. The typical Siemens bribe is 5-6% of the contract value. Let’s take 6% just to be on the safe side. According to this ratio, the Greek state must have paid between EUR1.5bn and EUR2.2bn per year on bribes. 

But of course bribery isn’t just about paying the actual bribe, it’s also about buying inferior services or paying over the odds. The bribe is meant to convince officials to allow this. These additional economic ‘capture’ costs come up to anything from 20% to 188% of the bribe itself.  Combining this calculation with the estimates on bribes it is possible to estimate an upper and lower bound for the cost of bribery and corruption in public tenders for Greece.

Now I realise that in applying these rules to all public tenders I am making a heroic assumption – some contracts will have been pimped to death, with contractors making incredible capture rents, and others will have been done by the book. I am also assuming that bribery and capture costs remained constant as a percentage of procurement spend every year, which can’t be true as there have been procurement bonanzas that will have been milked to death during this time, as well as some years when rents from bribery were low. 

I can’t help this miscalculation given the tools at my disposal. It’s just the best estimate I have. And it looks as follows. The total costs of capture (bribes and mispricing) ran up to anything from EUR900m to EUR5bn per year.

Now, in determining the extent to which these rents contributed to Greek Government debt, I must make some assumptions about their financing. To ensure I cannot be accused of bias I will make the most defaultnik-friendly assumptions possible, in the understanding that they may be biased in favour of overestimating the odiousness of the stock of Greek debt.

First, I will assume that all of this money came from the Greek public coffers. This is patently not true as EU money flooded into the country from 1995 to 2009 and much of it went towards procurement.

Then I will assume that all of this money came from excess borrowing and thus a) we are still saddled with the interest to this date and b) this debt is indeed odious. This is a very strong assumption and one that is moreover heavily biased towards the defaultnik case.

This means I need to calculate an acceptable interest rate for the excess borrowing. Given that Greece never paid down any debt but simply refinanced existing obligations throughout this period, I feel justified in calculating our effective interest rate by dividing the total stock of debt for each period with the total interest expenditure for each period. Both can be found here. I assume that costs before 2000 (when the Eurostat series begins) were constant at the same level as 2000, i.e. 7.2%. (Note: they were actually higher).

Now all that remains is to calculate compounding coefficients for each year based on the product of the (interest rate+1) for that year and all following years. They look as follows:

Now all that remains is to add up the up-to-date figures. The Grand Total comes up to a range of EUR29.8bn to EUR71.6bn, or alternatively 9.1% to 21.8% of our total stock of debt as of end 2010. 

Remember, these are very generous figures, and yet even on these assumptions, the amount of potentially odious debt is almost certainly less than the nominal 21% haircut agreed in July.

With procurement out of the way only straightforward graft and over-compensation of officials remain as possible avenues for the creation of odious debt. However, I believe that the contribution of these two is negligible compared to that of public procurement as indeed it is in almost any country not run by warlords.  

UPDATE: I realise in defending these estimates that there's just no pleasing some people. If you're not happy with my figures or my assumptions, let's at least agree on this: That it is possible, in theory if not in practice, to come up with a good estimate of the amount of debt attributable to things other than the will of the people; that carrying out such estimates is desirable; and that the extent to which Greece's debt is odious is a matter of fact, not politics. My assumptions are no doubt flawed but they are transparent, they come with some justification, and they are there for all to evaluate. In fact, you can just plug in your own assumptions and try to get an estimate that works for you.

2013 UPDATE: How fair is @talws' objection in the comments below? I explore the topic here.

2014 UPDATE: How accurate was my 6% assumption on the size of bribes, on which so much of this exercise depends? There is a new dataset for the researcher to draw upon: the recently released OECD estimates on the size of bribes as a % of contracts, based on records from 55 actual cases brought to justice between February 1999 and June 2014.

The average OECD estimate is 10.9%, which is significantly higher than my assumption. Bribes, of course, range widely by sector, from 14% for health- related spending (one of my Big Five deficit-drivers) and 17% for admin services, to 6% for scientific and technical consultancy, and 4% for construction. Ironically, the types of projects most commonly cited by defaultniks back in 2010 (Olympic construction for instance) attract relatively small bribes (less than my original estimate), while services- and consultancy-based projects are much worse. Anyway, please note these figures are based on a vanishingly small sample, involve foreign bribery only and do not seem to include any cases involving Greece. Still, if this estimate is accurate, then my estimate of our odious debt should grow by 81% to roughly 16.5% to 39.5% of Greece's 2010 debt. The mid-point of this range is now above our original creditors' haircut, but nowhere near the total debt jubilee defaultniks were after.

Once again, the facts simply don't bear out the Debtocracy story. They do reveal a good amount of debt we could have done without; and I wouldn't mind defaulting on that even now. But at least we would be seen as credible and honest, as opposed to opportunistic and hypocritical.


  1. 1. Α counterpoint: 70% of my personal income goes to eating, renting a house and basic expenses. 20% goes to gasoline, recreation etc and another 20% goes to gambling and 5% to binge-driniking.

    In order to sustain my personal 10% deficit I have reverted to loans and credit cars, which I obviously can't pay up anymore as they are by now increasing my deficit.

    According to the reasoning above, my main problem is that I eat too much and move to a cheaper flat. There is a fallacy here :-)

    2. Since you are using procurement numbers up to 2009, you should calculate the "odious debt" as a percentage of the debt in 2009 not 2010. There are other reasons to avoid including calculations of the extra debt in 2010 obviously

    3. Having 20% of GDP less debt *at the start of the crisis* would certainly have had an effect in the speed of development of the crisis, at least...

    Having said that, kudos, this is a really useful calculation no matter where you stand on the issue.

    1. The problem is that you are trying to calculate the percentage corresponding to the briding..
      Odious Debt proclaims that the whole amount of debt (bridery + usefull part) created by illegal means should be written off.. Not only the malicious part..


Please remember that I am not notified of any comments and will not respond via comments.

Try to keep your criticism constructive and if you don't like something, do tell me how to fix it. If I use any of your suggestions, you will be duly credited.

Although I'm happy to entertain criticism of myself in the comments section, I will not tolerate hate speech. You will be given a written warning and after that I will delete further offending comments.

I will also delete any comments that are clearly randomly generated by third parties for their own promotion.

Occasionally, your comments may land in the spam box, which may cause them to appear with a slight delay as I have to approve them myself.

Thanks in advance for your kind words... and your trolling, if you are so inclined.