RBS’s Kieren Davies, who is one of the best local economists going around, has built a neat little empirical model of long-term government bond rates as a function of a range of variables, such as short-term rates, offshore rates, inflation and debt. While I will not disclose the substance of the research here, one finding is worthwhile highlighting. Kieren’s model suggests that for every $10 billion increase in real public debt, long term government bond rates rise by circa 0.07 percentage points. This is not an enormous elasticity. To put it in context, it implies that if government debt were to increase by, say, $50 billion, interest rates would rise by 0.35 percentage points. Fwiw, the chart below shows the time-series change in these two variables since 1990. I believe Treasury has recently published a paper on this subject, but I have not yet got around to reading it...
