What have we learned from the A-К Model? The answer is some things that we expected and some things we did not. As expected, we confirmed that the choice of the tax structure, the progressivity of income tax rates, the expansion of pay-as-you-go social security benefits, and deficit-financed tax cuts can very substantially alter an economy’s long-run well being. We also confirmed, as Chamley’s (1981) seminal work had suggested would be the case, that consumption taxation was more efficient than either income taxation or wage taxation and that capital income taxation by itself was the least efficient of all four tax structures.
What we didn’t expect to find is that economic transitions are really quite slow. Policies that, say, reduce the economy’s capital stock by a third in the long run can have a half life of 20 to 30 years. We were also surprised to find that the duration of a policy matters a great deal to its short-run impact. Take deficit-financed tax cuts. If these are of short duration, substitution effects dominate income effects and the economy expands as workers take advantage of temporarily low tax rates.
The short-run expansion can be quite substantial, not so much because of the size of compensated substitution effects (our parameter values are fairly conservative), but because of the magnitude of the size of the changes in economic incentives.
So the supply side zealots in the U.S. are partly right; deficit-financed tax cuts can raise saving, crowd in capital, and expand output. What they get wrong is that these gains are short lived. Although short-term tax cuts lead to short-run crowding in, they lead to long-run crowding out and a lower long-run level of capital and per capita output than initially prevailed. Why? Because the economy’s tax base doesn’t expand in the short run by enough to permit a permanent reduction in tax rates.
When the end of the period of the tax cut is reached, the government must raise tax rates, above their pre-policy level, in order to cover its spending plus interest payments on the new debt it incurred up to that point in time. This was true in our simulations no matter what tax rates were being cut, including capital income tax rates.