As the $1.9 trillion Covid-19 relief legislation moves to the Senate, it temporarily distracts attention from even more ambitious legislation that’s set to follow—President Joe Biden’s Build Back Better recovery plan—and the battle it’s sure to provoke. At the heart of this brewing fight is the question of which does more to protect the future: addressing climate change or containing fiscal risks. Because it probably won’t be possible to accomplish both, climate mitigation must come first.
The recovery plan will include trillions of dollars of investment in infrastructure, clean energy and other things. And the coming debate will consider whether that new spending should be offset by revenue increases or spending cuts elsewhere in the budget.
The Biden administration has signaled that the permanent parts of the infrastructure and clean-energy package will indeed be paid for, rather than adding to the deficit, and this is what many moderate Democratic senators also favor. Thus, any long-term spending on electric-car charging stations, carbon capture, aviation, water infrastructure, broadband and other priorities in the bill will be tied to tax increases or reductions in other spending.
The idea here is to avoid worsening the country’s long-term deficit. However, as Robert Rubin, Joseph Stiglitz and I noted in a recent Peterson Institute for International Economics paper on fiscal policy, budget projections are subject to substantial uncertainty, and that complicates the question of when and where offsets should be imposed. Indeed, this issue—whether new programs that entail long-term changes need to be linked to deficit-reducing measures—was the most prominent area on which the three of us could not reach agreement. As we wrote:
Although we agree on much about how fiscal policy should be conducted, we disagree along several dimensions. As one illustration of the ambiguities inherent in an uncertain world, for example, the three of us have different perspectives on whether any spending increases or tax reductions enacted today but that extend past the end of 2022 should be offset by other changes in the budget. One view is that because it is politically easier to cut taxes and increase spending than to do the opposite—and because of the uncertainty about future fiscal constraints—new spending increases or tax cuts that extend beyond the end of 2022 should be offset with deficit reduction measures. Policymakers could then undo those offsets in the future if the economy remains weak. A second perspective is that any such offsets should be automatically triggered off if an economic indicator, like the unemployment rate, signals continued weakness in the future (so with similar effects but less discretion than the first perspective). A final perspective suggests that because a robust economy benefits lower-income groups the most, particularly those previously marginalized, one should wait for more compelling evidence that further fiscal stimulus is not needed before enacting the offsets.
That discussion applies to a generic program; it does not account for the lost opportunity if a budget-neutrality constraint is imposed and, as a result, climate legislation is either not enacted or substantially cut back. (The legislation may be enacted under budget reconciliation, which imposes specific rules on measures that expand deficits over the long term. Failing to offset the spending would then require ensuring that no such spending extends beyond the relevant budget window under reconciliation, if necessary by sunsetting the spending before the window closes.)
The core problem is that while deficit offsets are theoretically a good idea, they are very hard to legislate in practice. I vividly recall many times, when I was director of the Congressional Budget Office, being summoned to a senator’s office to discuss options to lower the deficit. I would bring my massive binder full of possibilities, and walk through them one by one—and would inevitably be told that each one was just not possible. Gradually, billions in savings would shrink to millions, and then perhaps hundreds of thousands.