12 August 2013

Savings and Interest Rates

Frances Coppola has been writing about what determines the interest that savers earn on their accounts. On Twitter she asked what do we mean by savers, people who save out of current income or people who hold savings accumulated over time. During the discussion we got into how interest rates affect decisions to save.

To me that is an empirical question. Do higher interest rates incentivise saving?  Do lower interest rates encourage people to save more to meet a savings target? Or is saving unaffected; people save what is left over after paying for everything else?

When I studied macro we had a diagram which showed the loanable funds theory of interest rates. (The interest rate is the price that matches the number of borrowers to the number of savers.) In this diagram the supply curve was vertical, meaning that the level of saving was independent of the interest rate. I always wondered if that really was the case. As I say it is an empirical question.

Using some data I have to hand (The BoE's useful three centuries of data series). I have done a scatter chart.
Source:BoE and others
I have used long run government bond yields as an index of interest rates and the savings rate in the UK from 1948 to 2009. This data is not ideal, but it gives a quick and dirty answer. Other countries and time periods might give a different result.

The answer is that the savings rate is higher when interest rates are higher. The R squared is 0.59 suggesting that interest rates explain 60% of the increase in savings. In the loanable funds model the supply curve would be upward sloping.

The one issue I am aware of is that I have not controlled for inflation. It would be better to use real interest rates, which I will do when I can look for another data series.


  1. If interest rates tend to be higher during a boom, could wages/spare cash be higher when rates are higher?

    (Sorry, completely lack the skills to test this.)

  2. I'm not sure. Higher incomes would mean a higher saving rate. On the other hand, the highest interest rates in this series occurred during the stagflation of the 70s. I doubt there is a stable relationship between nominal interest rates and growth rates, but I don't know.