Consumption Income Wealth and Expectations

I learned a fact from Chris Dillow who went on to conjecture that “consumers -in aggregage – have genuine foresight”

I agree that “confidence” is a non-explanation for fluctuations in consumer spending. Such fluctuations can, for the most part, be explained by observable economic variables such as incomes, unemployment and credit availability, as John Muellbauer, for example, has shown. Insofar as spending is forward-looking, it’s not because of confidence, but because consumers – in aggregate – have genuine foresight; this is why consumption-wealth ratios help predict equity returns.

I won’t pretend that I have an open mind on the question of whether there is any sign of genuine foresight in the economy and will just jump to my counter arguemnt (in a comment so addressed to him).

There is an equivocation on the part of the post about consumption. When discussing determinants of consumption other than expectations you list

“incomes, unemployment and credit availability,” When claiming that expectations matter and are close to rational (at least positively correlated with rational expectations) the other variables suddenly change to wealth “consumption-wealth ratios help predict equity returns.”

Clearly the analysis is invalid. You have to say that consumption relative to that expected given wealth, incomes, unemployment and credit availability is correlated with equity returns.

I personally have no doubt that the correlation you mention is due to the effect of irrationality on equity returns. In aggregate (and the cross section) low p-e ratios are correlated with high returns. Equity returns over 5 years periods are negatively correlated. Other things equal low equity prices are correlated with high returns. Other things equal low equity prices are correlated with low wealth. Consumption is smooth (whether because permanent income is smooth or because of habit formation) .

I think what you report is if you take two smooth series (a moving average of corporate profits and consumption) and then look at the ratio each to insanely bouncing equity prices to each you find the ratios are positively correlated so (consumption/wealth (including equities)) is correlated with smoothed earnings/equity prices which is known to be correlated with equity returns.

I have no doubt that your evidence for some rationality is based on extreme irrationality. In particular I know for a fact that a high ratio of consumption to disposable income is not correlated with high growth of disposable income. This is obviously a much more reasonably way to decide if there is anything to the PIH.

I think that it is clear that the PIH was a turn down a blind alley. The evidence you present has absolutely no effect on my beliefs, as it is obviously of little relevance to the question.

Comments (3) | |