A Whole VAT of Stimulus

Over at Atlantic Business, our own Derek Thompson is channeling Bruce Bartlett to suggest that announcing a VAT now, to take effect in a year or so, might serve as a nifty alternative to a stimulus.  After all, if a big tax hike is coming, you'd better stock up on pasta and paperweights.

This is not a crazy idea--I heard an economist who specializes in Japan suggest basically the same thing ten years ago to cure the "lost decade" (though his was actually a straight consumption tax that stepped up every year).  As Bartlett told Ezra Klein:

Suppose you had a 10 percent VAT and we said we weren't going to collect it for the next 10 months. People would buy like crazy. They'd buy toilet paper, they'd buy anything they could get their hands on that they knew they'd need in the future. We're depriving ourselves of a great stimulant tool by ignoring this.

It would also have the happy side effect of solving some of our worst budget problems--and no, conservatives, we're not going to make Social Security and Medicare go away, so we're going to have to explore some means of paying for them.

The biggest problem I see with this is that the VAT is, as Bartlett says elsewhere, too good a tax--it would be very hard for a government facing a massive revenue shortfall to push through a tax holiday, and once the tax was in place, it would be hard for the many links in the compliance chain to temporarily abate it. 

The idea behind a VAT is that every link in the supply chain pays tax on the value it adds to the product--to simplify greatly, the difference between the cost of its inputs, and the sale price of the product.  This is what makes the tax so powerful:  it's hard to dodge, because the next link on down the supply chain does your compliance work for you. Since they only want to pay tax on the value they add, rather than the total sale price of their product, each firm makes sure that the VAT was properly paid on the inputs in earlier stages.

But production takes time, and this complicates your tax holiday.  On day one of the tax holiday, does Home Depot credit you with the full value of the VAT on the new grill you buy?  And if so, where do they get the money?  Remember, they've already paid the accumulated VAT to their supplier (if they haven't, you don't have a VAT; you have a sales tax).  Similarly, after the holiday ends, you'll have a bunch of products that were partially produced during the holiday.  These sorts of compliance problems mean that a VAT holiday is a one-time trick:  you can announce a VAT and then delay the introduction, but once you've implemented a VAT, you can't really take temporary breaks--at least, not with the sort of transparent instant effect you want for a good stimulus.

The other problem is that I'm not sure how well this sort of stimulus works in the times when you most need it:  when consumers are credit constrained.  There are a lot of people right now who can't go out and stock up on 8 months worth of toilet paper (even assuming they have the space) because someone's lost their job, their credit lines are being slashed, and the family is being forced to live paycheck to paycheck.  Cutting their VAT will make their current dollar stretch a little further, but it is not going to encourage them to time-shift consumption.  As household finances improve, the stimulus works better, but also is less needed.  So something like a straight increase to unemployment benefits, which directly targets the most constrained households, may have better stimulative power, while a payroll tax holiday is probably easier to implement and has beneficial side-effects on employment.

That doesn't mean we shouldn't have a VAT (though I'm skeptical, mostly because they often are used to endlessly ratchet up the size of government).  I'm just not sold on the stimulative prospects for changes in the VAT.