Update No. 111 – 03/05/13

Both EGP & our benchmark closed at all time highs since the fund was founded today, given we don't own any bank shares, leading the Australian benchmark quite handily this year is a pretty good accomplishment I think…

I commented on the ‘Top 5’ stories from my Twitter feed a couple of weeks back. I realised when writing that there was one other heavily discussed issue that week, but I wanted to poke around it in a little more depth.

It is not particularly relevant to markets in the short-term, but is very much of interest to persons (like me) who are interested in the minutiae of economic policy.  The issue was major blow-up in a widely loathed and respected (in approximately equal parts) academic paper by Reinhart & Rogoff.  The issue was to do with an error in an excel formula, which is pretty embarrassing, but the authors insist it doesn’t change the thrust of the paper.

The paper it should be noted was controversial before the error came to light, but the controversy was primarily centred around the authors suggestion/implication that a national debt/GDP ratio of around 90% is something of an ‘inflection point’ in a nation’s economic fortunes. Those who disagreed with this idea jumped on the ‘excel error’ to completely rubbish any suggestion that there could be anything of merit in the paper. When I look at the graphs below, I’m very much inclined to agree with those who say the 90% point is all but meaningless, but it seems very clear that real GDP growth is strongly negatively correlated with national debt levels and that at higher national debt levels, the relative harm to RGDP growth increases. If partisanship could be put aside, there is valuable information to be gleaned from the thrust of the study.

The best analysis came here, and from that blog I borrow the following graph:

The dotted lines are 95% confidence bands. The author commented as follows:

There is a visible negative relationship between growth and debt-to-GDP, but as HAP point out, the strength of the relationship is actually much stronger at low ratios of debt-to-GDP.  This makes us worry about the causal mechanism.

This statement makes sense when you first consider the graph – the confidence bands do become meaningfully wider as debt/GDP ratios grow. When you pause to consider why that might be though, I think it makes more sense and we can still assume some level of causality – if there are some countries where it is driven more demographic factors, then mitigations should be put in place prior as such drivers/factors should be visible years prior.

The vast majority of countries operate the majority of the time at the lower end of the scale, say between about 25% and 100% debt/GDP. At these levels,  there are plentiful data and the experiences should be expected to fall into a narrower band due to this region being the most ‘normal’ economic area. When you get to either end, however, the bands widen out markedly.  I expect it is because of the small sample sizes of the groups very close to either 0% debt/GDP or greater than about 150% debt/GDP ratio and the fact that at these levels the economic actors within (and outside) these economies are not always confident how things will play out, so the range of expected outcomes widens.

This is simple ‘expectation economics’ where if enough people expect a certain outcome, that outcome is more likely than not to be delivered. Rational expectations in the 30-80% debt/GDP range seems to be for about 3% RGDP growth, so that’s what you ‘normally’ get. When you get out near 200% debt/GDP, actors don’t have a lot of historical evidence to based their expectations on and so you get a wider range of outcomes. What is pretty clear though is that the trend is for lower RGDP expectations when debt/GDP gets to high levels. There is almost certainly some corellation in this, but to rule causation out, in an ‘out-of-hand’ manner is a huge mistake I think.

I think most people understand instinctively that lower real growth has to go hand-in-hand with higher debt as more of your outputs are siphoned off to service debt. As debts approach levels where ‘serviceability’ starts to be a consideration, RGDP will fall as the possibility of borrowing for new expenditure is reduced.

Countries with near zero debt/GDP ratios have considerable borrowing power at their command, which can be used to focus on infrastructure needs that will provide a meaningful RGDP boost. At the other (say around 200%) end of the range, even if there was a good project by which future RGDP could be with some certainty increased, it may be rejected out of necessity.

Though I can understand why people have their partisan views, based on their political preferences, it seems to me that information such as that graphically presented above should clearly present a persuasive argument for aggressively paying down debt (n.b. ratio levels – this doesn’t necessarily require running a surplus) during times of normal economic output. When a country finds itself in the 0 – 50% range, options when confronting any downturns are greatly enhanced.

Persons with a predisposition to ‘like’ expenditure, who are comfortable with debt say ‘a country is not a household’ (at least when the Government they prefer is in office…). This is clearly true, but presented with graphical information such as that above even the most deficit loving partisan should be able to accept that less debt is broadly better in normal economic times.

The graphed line above would seem to indicate that any nation with about 250% debt/GDP will probably have a RGDP growth expectation of pretty close to 0%. Although this is (thankfully) a rarely tested proposition – Tony Hansen 03/05/13

EGP Fund No. 1 Pty Ltd. Up by 17.98%, leading the benchmark by 6.25% since January 1^st. Since inception, EGP Fund No. 1 Pty Ltd is Up by 43.61%, leading the benchmark by 27.17% all-time (April 1^st 2011).