Debt and gilts: some history

By Chris Dillow, 12 March 2010

Is high government debt a reason to expect gilt yields to rise? The message from UK history is clear - no.

My chart shows this. It compares the ratio of government debt to GDP to the yield on 2.5 per cent consols, a long-running undated gilt. You can see that there is no correlation between the two; statistically speaking, it was actually negative (minus 0.26) over the 1870-2009 period. In the 1920s, the debt-GDP ratio averaged 164 per cent - twice its current level - but gilt yields averaged 4.6 per cent, less than they yield now. In the 1950s, debt averaged 144 per cent of GDP and gilts yielded 4.3 per cent. The worst rises in gilt yields - in the late 60s and 1970s - actually came as the debt-GDP ratio fell sharply.

History, then, shows that government debt doesn't raise gilt yields. Sadly, though, this does not mean we can dismiss fears of a debt crisis. There are several reasons why the gilt market wasn't spooked by the massive debts left by the two world wars - and all these reasons might be weaker now than then.

First, and most importantly, inflation then was low and was expected to remain so. Yes, gilt yields rose after WWI as inflation soared. But they fell back in the 1920s as deflation - a common experience in the years before the war - returned. And inflation, traditionally, has been the key determinant of gilt yields. Over the 1870-2009 period, the correlation between the consols yield and current inflation was 0.54 - amazingly high, given that it is inflation expectations rather than current inflation that should matter for yields.

Herein, perhaps, lies one difference between today and the two post-war periods. Back then, no-one expected sustained inflation, simply because it did not happen in peacetime; in 1914 the price level was lower than it was in 1870, and 1939's prices were lower than 1922's. Investors were therefore happy to buy gilts yielding less than 4.5 per cent, in the belief that money would not lose its value over time. Today, though, almost everyone expects future cashflows to be devalued. The gilt market expects the price level to rise 60 per cent over the next 14 years. And with some commentators calling for a rise in the inflation target, it's possible - though not certain - that it will go even higher than this.

Secondly, in the two post-war periods confidence in gilts was under-pinned by huge cuts in government spending as military spending fell. Between 1918 and 1924 the share of government current spending in GDP fell from 39.9 to 17 per cent, and from 1945 to 1951 it fell from 52.8 to 27.4 per cent*. The ratio is 43 per cent this year, and the CBI is asking that it fall by 8 percentage points by 2015-16 - far less than the post-war falls.

Thirdly, after both wars gilts were regarded as the natural home for investors' money, because the "cult of the equity" had not yet started. Investors didn't appreciate - to the extent we do now - that gilts are risky because of the danger of inflation, or that equity risk can be partly diversified away or that decent equity portfolio will, over time, grow as GDP grows. All this meant there was a big demand for gilts - relative to the size of the economy - that doesn't exist today.

A fourth difference between now and the 1920s and 40s is that back then investors had decades of experience which told them that gilt yields would stay in the 2.5-5.3 per cent range. This, therefore, was where they expected them to stay. "The rate of interest is a highly conventional phenomenon" wrote Maynard Keynes. "Any level of interest which is accepted with sufficient conviction as likely to be durable will be durable."

Today, however, things are different. Even your youthful correspondent remembers that gilt yields have been above 10 per cent during his working lifetime. This means that when we hear pessimistic talk that gilt yields could rise to 7 or 8 per cent, we instinctively think it at least a little plausible because we remember them being at that level. The psychological anchor for gilt yields is much weaker now than it was in the 1940s.

You might think there's another difference between now and the 1920s or 40s. Compared to then - though not to the pre-World War I period - we now live in a globalized economy.

But this is a two-edged sword. On the one hand, it means there are more substitutes for gilts, so if investors do worry about UK government debt they are more able to sell gilts and buy adequate alternatives. But on the other hand, the government today can tap the pool of global savings, whereas in the 1940s it was more dependent upon purely UK savings - a fact which greatly helps hold yields down.

The message of all this is that - for once! - the cliche is right. Things are different now. The factors that allowed previous governments to run up huge debts whilst nominal gilt yields stayed low are weaker these days.

This is not to say that we are heading for a debt crisis in which gilt yields rise sharply. It just means that, whatever reasons there are to believe there won't be a crisis, history is not one of them.

* On both occasions, though, the share of spending in GDP was much higher five years after the war than it was in the year before the war. There was a strong "ratchet effect" in the growth of the state.

The figures. The debt-GDP ratio is taken from Christopher Chantrill's superb site, ukpublicspending.co.uk. Consols yields come from Friedman and Schwartz, Monetary Trends in the US and UK for 1870 to 1966, and from National Statistics thereafter. Inflation figures come from National Statistics, and the share of government spending in GDP from C.H. Feinstein, Statistical Tables of National Income, Expenditure and Output 1855-1965.

visible-status-Public story-url-debtgilt_markets_190310.xml