A Global Savings Glut? Or Just Deflationary Forces?

One of the reasons that we tend to eschew “black box” forecasting models of any economy (aside from the fact that our econometrics courses now seem to have been a long time ago) is that we suspect that there are simply too many variables and discontinuities for even mathematicians with the skill of the late John Nash to ever really encompass effectively.

In this context, one of the least understood or modelling-friendly “variables” within a macroeconomic system is the household savings rate. For example, overall demographics within a country will play an important role in determining aggregate savings behaviour by the population as a whole, while interest rates and interest rate expectations will also exert an important, but probably non-linear, influence on people’s savings behaviour. We suspect that the relationship is essentially “U-shaped” in that people will save more either at very low yields or very high yields – the “bottom” of the savings rate will likely be somewhere around the 3-8% yield range (in the US and elsewhere). 

The absolute level of incomes will also impact the savings rate but so too will the distribution of incomes. If the income data is too skewed with a lot of less well-off people and a lot of very high earners, many people may simply not be able to afford to save. Asset prices and “wealth effects” will also play some potential role in determining the amount of current income that is devoted to savings, although the coefficient on this particular factor will likely vary according to whether the higher asset prices can or cannot be used as collateral within a fully functioning (or not) credit system. In particular, the availability or otherwise of credit will play a large role in determining the savings rates, as will any legacy effects of previous credit excesses. Similarly, household expectations over the outlook for inflation or deflation will also play some role in determining the savings rate, although we doubt that this term is quite as important as many analysts suspect.

Finally, we would also note that people’s expectations over the economy’s – and more particularly their own – future income trends will play a large role in determining just how much money people will wish to save in the current period.