Levies, borrowing, and spending cuts

The New Zealand government's looking for ways to pay for the rebuilding of Christchurch.

The Greens have pushed for an earthquake levy: 1.5% on income from $48K to $70K, and 3% on income above 70K. A few opeds have lent support to the levy. The Government has looked instead to a combination of borrowing and spending cuts.

Some folks argue that a levy is no different from spending cuts in terms of economic effects - both pull money away from people and reduce aggregate demand. I don't think that's quite right, but let's work it through.

First off, recall that income taxes come with deadweight costs: Treasury says that they're on the order of twenty percent in New Zealand. So raising a dollar's worth of extra income costs the country $1.20.

Let's assume that the ex ante fiscal position was optimal. We had the mix of spending that was best and the best mix of taxes to pay for it. The marginal dollar spent by government brought $1.20 in benefits in this the best of all possible worlds - an additional dollar raised in taxes would have cost more than it was worth. I'm pretty sure that's wrong, but we'll stick with it for now. In this world, we're indifferent between cutting a dollar's worth of spending and raising an extra dollar in taxes: both cost $1.20.

A big honking earthquake destroys a good chunk of Christchurch. The marginal benefit of an additional dollar's spending increased, but not because existing programmes have become all that much better; rather, it's because spending on rebuilding Christchurch infrastructure is a pretty high value investment. Spending right now on preventing the sewerage system from collapsing into a cesspool and miasma of poo-gas is worth an awful lot more than what we're paying for it. So we want to rebalance the composition of spending until the marginal value of a dollar spent on rebuilding Christchurch is equal to the marginal value of a dollar spent doing other stuff. Optimal policy will have to involve transfers within the portfolio of government spending. And so there will have to be cuts to other programmes.

Aha you say! But after this rebalancing, we'll want to ensure that the marginal value of a spent tax dollar is equal to the cost of raising it; we'd then want to increase the overall size of government. But recall that the deadweight costs of taxation are increasing in the tax rate: the Laffer curve isn't just about the point at which tax increases become self-defeating; it also reminds us that even well prior to that point, the economic costs of higher taxes are increasing in the overall tax take. Equilibrium would then involve an increase in the overall size of government, but still with some cuts to ex ante marginal projects because the deadweight costs of taxation are higher at the larger overall tax burden. That's shifts along a deadweight costs curve. And, it seems rather likely that the deadweight costs of taxation have increased as consequence of the quake - the curve itself has likely shifted. If you're targeting tax hikes on high income folks, you've gotta remember that those are the ones who are internationally mobile and, if they live in Christchurch, are probably already weighing up exit options. If labour becomes more elastic because folks are happier about leaving, the deadweight costs of taxing them has also gone up. Maybe you could build some argument about patriotic fervour reducing the deadweight costs of taxation post quake, but I wouldn't believe it; if anything, it's more an argument for things like war bonds for reconstruction paying relatively low interest rates.

Finally, we have to think about the optimal mix of tax increases and debt. If the prior mix of spending and taxes were optimal, then we'd likely be looking for a big increase in debt now to be paid off by a mix of future tax increases and cuts to non-earthquake programmes. Why future tax increases rather than current? The quake was a massive negative productivity shock for the region, so it's just more expensive to raise a dollar's worth of taxes today than it will be in four years. It makes sense to take on debt during a negative productivity shock to pay off later on when the production function's gotten back towards normal.

So even if you reckoned the ex ante mix of spending and taxation was optimal, the best response to the quake seems likely to involve cuts to non-quake programmes, increased debt, and future tax increases. If you figure, like the Greens, that the ex ante mix had way too little government spending, you'd push for smaller cuts to non-quake spending and larger future tax increases; it's beyond me why they're averse to debt to fund a massive capital reconstruction project. If you figure, like National, that the ex ante mix had too much government spending, you'll bring forward some of the cuts that you'd already had in mind while raising debt and pushing back future plans for tax reduction. If you reckoned, like me, that a reasonable proportion of government spending was of negative value, the case for cutting that spending has gotten stronger.

It's hard to avoid that the optimal arrangement involves some cuts to relatively low value government programs, even if you think those programs are of positive value.