Rethinking Economics: secular stagnation

by George Hatjoullis

There is a popular idea in circulation referred to as ‘secular stagnation’ and widely attributed to Lawrence Summers. The idea it seems is based on the notion of a chronic excess of savings over investment. The result is an economy operating well below potential. The weakness in such ideas is that they never really explain why savings should equal investment beyond some vague notion of the equilibrium real interest. This simply will not do. To elaborate I need to refer to my schema of the Economics series 1-19.

Recall that GDP is the economics world is method of adding apples and pears in terms of a fruit basket. The precise composition of the fruit basket is secondary as the basic measure is the value of the basket in a given year (nominal GDP). Comparison across time is achieved by revaluing each year’s fruit basket in terms of the prices prevailing in some arbitrary year (base year). Real GDP growth is measured by the change in the value of the series of constant price fruit baskets. From these two series we can derive a measure of aggregated price changes known as the GDP deflator. Nominal GDP in any one year accrues in the first instance to the owners of the means of production and the labour employed to produce. It is redistributed via taxes and subsidies (government) to different parts of the population and some is exchanged for foreign produced GDP (net exports). Those that have claim on it need not use it themselves but may exchange it for some financial claim that can be redeemed against future nominal GDP (saving). Many financial claims have already accumulated (wealth) in the hands of economic agents and can be redeemed in any one year (dis-saving). Those not having a claim can borrow nominal GDP in any year by issuing a financial claim to someone who has a claim and wishes to save (dis-saving). Some nominal GDP is deployed to produce more GDP in future years (investment). This schema usually comes with algebraic notation adding an air of pseudo-scientific precision and should be familiar to all economics students at all levels.

The question posed is what ensures that the nominal GDP that savers do not want to use in any one year equals the amount wanted by dis-savers and investors? The economists answer is usually phrased in terms of the rate of time preference aka the interest rate. Lower interest rates discourage saving and encourage dis-saving and investment. So why can we have a situation such as secular stagnation? The simplistic answer is that the nominal interest rate cannot fall below zero and the rate of interest that equates the two is negative (I am deliberately ignoring inflation). There are two serious problems with this specification. First, empirical. Negative interest rates and bond yields are quite evident at the moment. Second, it does not explain the why there is a secular excess of desired savings over investment. It is possible to formulate theories of secular stagnation based on inequality of income distribution and technological change and there is considerable merit to these theories. However, it also possible that what we are observing is much simpler and the misunderstanding arises out of the oversimplified macroeconomic framework beloved of economists.

All savings are not equal. To keep it simple consider three classes of claim on future nominal GDP; risk free real (RFR), risk free nominal (RFN) and risky nominal (RN). RFR provides safe claim on future real GDP. RFN provides  safe claim on future nominal GDP. RN provides an uncertain claim on future nominal GDP, though presumably offers a greater return as compensation. RFN has some inflation risk and may command a risk premium over RFR. Investment is inherently risky so in order to finance investment savers must desire to hold the appropriate value of investment claims issued by those undertaking investment. If savers desire a combination of risk free and risky claims on the future the question arises why should savings ever equal investment? The latter does not create any risk free assets. Aggregate savings always exceed investment.

The excess of saving is taken up by dis-savers. Pensioners run down accumulated claims which may be well be RFR and RFN. Governments borrow to spend, issuing RFN and RFR claims. The same question still applies; what makes desired saving equal desired did-saving and investment? However, we now have another dimension; risk. It is no longer simply time preference that equilibrates the system but also attitudes to risk. What happens if savers want to own more RFR and RFN assets than are available? Recall that savers include owners of the means of production that might also be expected to undertake risky investments. There has been no increase in the aggregate savings rate but rather a preference for holding savings as risk free assets. Desired aggregate savings may well equal desired dis-saving and investment but risk preference has changed. The demand for RN has fallen relative to RFR and RFN.

The equilibrium economist will respond that the prospective return of risk free assets will fall and that of risky assets will rise until the two are brought into balance. Perhaps but note we now have a completely different interpretation on the secular stagnation phenomenon. It is not excess savings but an excess demand for risk free assets that is the source of the problem. This is what I have referred to in innumerable blogs as the ‘The Great Deflation” and it is quite different from secular stagnation as defined by Summers. It is a direct consequence of the 2008 financial crisis, of the eurozone crisis, of eurozone bank reform, of the Cyprus crisis, of austerity and bank deleveraging. It is not about to dissipate any time soon because the excess demand for risk free assets cannot clear in the present environment.