A puzzling proxy from The Economist

From The Economist:

[R]eal interest rates should be roughly the same as the trend rate of GDP growth (a proxy for the return on capital).

This isn't the first time I've seen this statement used there, and I once even wrote a letter asking about it. I've tried playing around with several expressions that describe the link between growth rates and interest rates, but I can't come up with anything that would lead me to think that the two would be the same, even approximately. For example, growth would be zero in a steady state, but that doesn't mean that real interest rates would be.

Am I missing something?

Comments

[Chris]

Beats me. An economy with zero growth (ancient Egypt?) might still have positive interest rates, just because people prefer consumption in the present to the future, and lending is risking. Although if saving were risking you might even have negative rates, but that's an unlikely story.

[John Quiggin]

One obvious implication of this equality is that, with balance on the trade account, the current account deficit, expressed as a proportion of GDP will be stable.

For a closed economy, I think you get this equality with an intertemporal elasticity of substitution equal to 1, which is plausible (don't quote me on this).

[knzn]

In the Solow model, golden rule growth in the steady state implies that the marginal product of capital equals the growth rate (and the interest rate equals the marginal product of capital). To the extent that golden rule growth is optimal and savers are rational with infinite horizons, they should choose their saving rate to produce golden rule growth, which would imply that the interest rate equals the growth rate in the steady state. I think, in some models with infinitely-lived optimizing agents, you do get this result, but as I recall, there are complications when you introduce things like depreciation and time preference.

[true dough]

Write them another letter, Prof Gordon! I've just read about this in Romer's advanced macro text. Romer comes to your same conclusion, saying the U.S. and other major economies have proven to be dynamically inefficient by the golden rule standard (IOW, the real interest rate < growth rate of economy). Interesting factoid: From 1926-1986 “the interest rate averaged only a few tenths of a percent, much less than the average growth rate...”
One problem is that the golden rule fails to stand up against uncertainty.
Romer points out that Abel, Mankiw, Summers, and Zeckhauser (1989) offered an alternative method to measure efficiency WITH uncertainty: “...A sufficient condition for dynamic efficiency is that net capital income exceed investment.” Following this method, Abel et al find that the US other major countries are efficient, including years 1929-1985 in the US.
Still, it seems weird that anyone would echo the golden rule sans raison d'etre.

[true dough]

My last sentence was sloppy. I just meant, it seems odd that The Economist wouldn't make a distinction between principle and practice.

[knzn]

Once you introduce uncertainty, though, you run up against the Mehra-Prescott paradox. IIRC, infinitely-lived expected utility maximizers would not require such high returns on capital, even with uncertainty, so maybe we are dynamically inefficient after all.