Fiscal rules can theoretically improve policy by eliminating “time inconsistency” among lawmakers. But the proposed fiscal trigger being discussed in the Senate tax reform bill would be a terrible fiscal rule.

You can see the thinking. Several senators worry about tax cuts blowing a big hole in the public finances. If they do not have the desired impact on economic growth, resulting in less revenue than expected, the budget deficit will grow and drive the national debt even higher. Concerned senators therefore seek a mechanism whereby if revenues are lower than expected, tax cuts will be partially reversed.

It’s welcome that some senators take the US federal government’s burgeoning national debt seriously. But there are obvious flaws with this plan (though we do not know details as yet), some of which have been discussed widely already.

First, how exactly will deviations in revenues be judged? An economy is a complex organism, and it is difficult to disentangle how much any change in revenue relative to forecasts is due to changes in tax rates as against other factors. Just look at the debate in the UK. Last week, the Daily Mail newspaper published a report that tax receipts following corporate tax cuts there had been much higher than expected. But experts from the Institute for Fiscal Studies pointed out that much of this was due to a faster recovery of corporate profits in the financial sector and the effects of Brexit, which had little to do with the changing rate. Under a trigger which merely judged revenues against forecasts, a host of things that affect revenues (both upwards and downwards) could be chalked up as the effects of tax policy, potentially resulting in damaging tax rises.

Then, as J.D. Foster notes, there are likely to be other tax policy changes and changes in growth forecasts in future years too. How will these be disentangled and the effects of this specific Act isolated? Will the trigger apply to just a particular revenue stream, such as corporate income tax revenues, or more broadly to capture all the spillovers of any investment boost? If the former, the probability that the trigger will be activated is highly dependent on the accuracy of any analysis of the incentives to incorporate versus operating as a passthrough. In other words, there are huge unknowns here.

Second, the inclusion of a trigger mechanism actually dampens the pro-growth effects of the tax plan, and risks lower-than-expected revenues becoming a self-fulfilling prophecy. Take corporate rates. On the margin, uncertainty about what the corporate rate might be in the long term deters investments today. Less investment today results in lower GDP and lower tax revenues elsewhere in the code. This lower-than-expected tax revenue then activates the trigger (if it applies across total revenues) which raises corporate taxes. There is good reason why economists say that tax policy should provide certainty and permanence in regard to rates. The GOP plan already has a lot of phase outs resulting from the Senate reconciliation rules. The last thing it needs is the risk of more.

Third, you do not need to be a Keynesian to recognize that an unforeseen recession, which would dampen revenues relative to forecast, would be a terrible time to worsen supply-side incentives by increasing the corporate income tax or marginal income tax rates. In truth, Congress would likely override the trigger in such circumstances. But if the trigger would simply be abandoned when it bound, then it suggests it is not a very well-designed trigger! Of course, there could be a recession escape clause, but similar logic applies more broadly if the economy grows more slowly than expected, due to reasons other than tax reform changes.

In short, a fiscal trigger that threatened higher taxes would introduce considerable uncertainty, risk tax hikes at the worst possible time, and could risk tax hikes when other factors resulted in lower revenue growth.