Temple’s paper seems to have more direct conflict, but it’s hard to tell from the abstract which doesn’t say much about exactly what he did. It would be surprising if the Solow model were the last word in growth, since it doesn’t actually explain long run growth. However, that doesn’t mean that its prediction of conditional convergence is incorrect and MRW is wrong. I was under the impression that the inability of a number of endogenous growth models to replicate the conditional convergence prediction was a stumbling block for many of them. Is that vaguely true?

And his most famous paper, “A Contribution to the Empirics of Economic Growth” of 1992, with David Romer, of the University of California at Berkeley, and David Weil, of Brown University, is generally considered to have failed in its defense of the “augmented Solow model,” meaning one in which human capital as well as capital and labor are sufficient to explain variations in growth.

The source had links.

I certainly wouldn’t call growth economics an empirical failure though. I think that the conditional convergence result of Mankiw, Romer and Weil is fairly convincing. It provides substantial evidence that the long-run predictions of exogenous growth theory are more than just fluff. Endogenous growth theory isn’t there yet, but it’s still a fairly young field and there are some very smart people working on it, so I wouldn’t want to give up on it.

Can we just say that growth theory is basically an abysmal failure empirically, and we don’t really have any results at all that one would stake one’s life on? 🙂 I’d still be more optimistic about a country with a growing population than a shrinking one.

I realize you have just given me the first google hit you found, but I wouldn’t say “significantly negative” when the regression cited isn’t actually statistically significant until the decision is made that the USA is an outlier and Luxembourg isn’t. If threw out Luxembourg and whatever the poorest country in the OECD is (?) you’d get a pretty flat line.

Checking the correlation you cited between fertility and gdp/capita, again I don’t know if I see much a relationship with incomes over 15k, and we’re still got the problem of causality to deal with.

I think the Malthusian argument alone probably explains most of the relationship we see in the total sample.

There could well be positive effects of the type you mention (although endogenous growth theory has a worse empirical record than the Solow-Swan model I referenced). However, they tend to go to the level of GDP, not the per-capita level.

Population vs GDP/capita isn’t important for growth theory but, interestingly, it is significantly negative in OECD countries anyway 😉

http://micpohling.wordpress.com/2007/08/18/oecd-population-vs-gdp-gini/

Stagnant population is probably bad (even though it implies rising capital/labour ratios), mainly because it implies problems for pay-as-you-go transfer systems. Is rising population in NZ good? I’ll be optimistic and say yes, citing network effects, thresholds, economies of scale, and the fact that NZ seems quite underpopulated to begin with.

go figure