Taxes and the Keynesian years 1981-2007

Thomas Colignatus 
December 11 2011 

The Reagan revolution since 1981 was not only supply-side but the deregulation of financial markets also released capital that flowed towards investments, which is a "Keynesian" demand-side effect. Disentangling these effects is not easy but still important in a recent discussion on the tax on the top 1%.

The tax cuts and deregulation started by President Reagan in 1981 and continued till the credit crunch of 2007+ are commonly seen as "supply side" but also contain a strong demand or "Keynesian" side. Obviously, tax cuts create spending power. Less obviously, deregulation notably in the financial sector liberated huge flows of capital looking for profit. This was a stimulus of non-fiscal and non-monetary nature but a stimulus nevertheless, and accomodated by the Central Banks rather than neutralised. It created the investments notably in the ict sector, the dotcom industry and the financial sector itself again, while cheap money also fueled debt-based consumption. Policy measures thus have effects along different channels, and only partial effects might be classified as purely supply or demand. The main thrust of the Reagan revolution thus lies not on the supply/demand distinction but on fundamental goals, such as on more income inequality and less parliamentary control of social-economic processes. To evaluate its success, it is necessary to correct for the "Keynesian" demand-side stimulus, otherwise we could exaggerate.

Piketty et al. (2011) consider the tax rate of the top 1% of the pre- and post-Reagan situations, and try to determine the impact. On income inequality they conclude: "the evolution of top tax rates is a good predictor of changes in pre-tax income concentration." On wealth creation itself: "there is no correlation between cuts in top tax rates and average annual real GDP-per-capita growth since the 1970s."

When considering two variables it is always wise to plot them and look at the correlation. But there can be confounders as well. There need not be much discussion about causality since the tax rate is a clear policy variable. A rational expectation might have been that a tax cut would cause deficits and debts that would cause a crash such that taxes would have to be raised again. A tax cut still leaves you with funds to invest and even Paris Hilton seems to have limits to her conspicuous consumption. The point is that the following can be confounding: (a) the cut in the top tax rate may have had a negative impact on growth, since a normal process of growth would have been social investments financed by such tax, as for example still happens in Scandinavia, (b) the source of growth came about by the "Keynesian" effect of the parallel policy measure of deregulation. This argument is based purely upon logic and has not been submitted to statistical analysis yet. 

In the mixed economy we require both private and public investments, with a dynamic balance between liberty and other goals, and if we do not account for above possible confounding then we risk that balance. Reregulation of financial markets will have a contractionary effect, and we should not neglect the need for national investment banks. The American stimulus of the economy in 1981-2007 has created a bloated financial sector and there are better alternatives.

Another complication lies with dynamics itself. Economics is about optimality and thus we look at marginal values and marginal tax rates. The usual marginal tax rate is the statutory rate, but there is also the dynamic marginal rate that arises from changes over time. Piketty et al. (2011) consider the statutory rate but my analysis in Colignatus (2011a) points to the importance of the dynamic rate. For example, if you have a piece of a larger cake, and if both piece and cake grow at the same rate, then the share in the cake remains the same. A person who pays 40% tax on average, when exemption is high and with a statutory rate of 60%, may continue to pay only 40% when exemption rises over time as well. In regressions on the Phillipscurve the average rate appeared to be much more important than the statutory marginal. See the Appendix for a formula.

Pikkety et al. (2011) have thrown new light on the impact of the Reagan revolution. I am a bit afraid that the discussion will be about their elasticities while the real clarification rather lies in the logic of the analysis. In my impression it is possible to already resolve current unemployment, see Colignatus (2011b).


Consider taxes T and income Y and the difference operator ? to calculate the changes of taxes and income. The rule is that (?T / T = ?Y / Y) <=> (?T / ?Y = T / Y), which means that if taxes grow as fast as income, with ?T / T the rate of change, then the dynamic marginal tax rate ?T / ?Y is the same as the average tax T / Y , and conversely. This formula also shows that it should not be a problem to have a high exemption of taxes and premiums at the level of the net minimum wage, such that labour costs can be reduced while maintaining a decent level of income. See Colignatus (2011a) for the use of derivatives.


Colignatus, T. (2011a), "Definition & Reality in the General Theory of Political Economy", 3rd edition, Thomas Cool Consultancy & Econometrics, see also

Colignatus, Thomas (2011b), "High Noon at the EU corral. An economic plan for Europe, September 2011",

Piketty, T., E. Saez, and S. Stantcheva (2011), "Taxing the 1%: Why the top tax rate could be over 80%",