A Primer on Ricardian Equivalence (or "Why You Should Put Your Tax Rebate In A Savings Account")...

(Note: thanks to Tim Harford for reminding me that this issue was bugging me.)

Ok, I've seen two tax rebates in recent years (one back in 2001 and one this year), and I get that the political idea is to get cash back into the hands of citizens so that they can use it to buy stuff. In economic terms, spending the tax rebate stimulates aggregate demand and (temporarily) raises the nation's output and income. (If you don't see this, consider that people wanting to buy more stuff creates jobs, and those people working the new jobs get money, which they use at least some of to buy more stuff, which creates more jobs, and so on. Look up Keynes and the multiplier effect if you're still curious.)

In a way, the government is hoping that people don't stop to think where the "free" money comes from. Where *does* it come from? Generally speaking, the U.S. government is not a profitable enterprise, so if it wants to give out money, it has to get that money from somewhere. (Given that there is a national debt, the money isn't just sitting around in a government vault.) There are basically two options- the government can reduce spending and give the money that it would have spent back to people, or it can borrow money to give back to people. Given that reducing spending is likely to lower aggregate demand, take some jobs away, etc., that doesn't seem like the way to go to stimulate the economy. (Keynesian economists argue that FDR's increasing goverment spending went a long way to get the U.S. out of the Great Depression, for example.) Therefore, the funding of the tax rebate is through government borrowing.

Why does this matter? Well, there is the simple matter that borrowed money has to be paid back eventually, with interest even. (Macroeconomic models typically rule out the possibility of a Ponzi Scheme that would allow the government to keep borrowing more and more money without limit to pay the interest on the existing debt. This is most likely a realistic limitation, though it seems sometimes as though we are trending in that direction.) Where is the government going to get this money? From future taxes of course! Essentially what the government is doing is borrowing money on your behalf (albeit at a lower rate than you could borrow) and giving it to you as a "rebate", but there will come a time when it needs the money back (plus interest) to pay its creditors (hello China). When you think about it this way, are you going to spend or save your borrowed money? I might spend it if I thought I was old enough to not be likely to see the future tax increase, but that's about it. Even then, I should probably leave it to my (hypothetical) next of kin, since they are going to have to pay for the rebate that I got.

Economists like to put names to theories, and they like even more to name theories after themselves. The idea that debt-financed tax cuts or rebates are ineffectual in stimulating demand because people realize that the money will be paid back later in the form of higher taxes is a variant on Ricardian Equivalence. (Note that the theory of Ricardian Equivalence would also imply that increases in government spending via borrowing would also be demand neutral, in contrast to the multiplier effect described above.) Clearly Ricardian Equivalance can't hold in practice in the tax rebate case, since if it did the government (hopefully) wouldn't bother with the rebate.

Thinking from the consumer perspective, when might a household choose to spend the rebate? Maybe, as noted above, the household unit is old and doesn't really care about its children (or doesn't have any). Perhaps the household realizes the interest effect, but also realizes that the government's interest rate is pretty low and decides that consumption now is worth paying the interest later, even if it wouldn't have chosen to borrow on its own in order to finance consumption. It could even be the case that the household realizes that its tax position changes over time (due to expected income and policy changes) and it's not likely to bear the brunt of paying back the rebate. (Those nearing retirement or planning a career change to street performer could be in this category.) All of these explanations seem perfectly rational, even in the strict economic sense, and are thus not worrisome to me. That said, there are explanations for rebate consumption that are likely not utility-maximizing in the long run, and they are the ones that make me feel like the government is taking advantage of human behavior a little. My list so far is the following:

1. Households are impatient and tend to overvalue present consumption and immediate gratification. Furthermore, households are often time-inconsistent in this way, which means that there will be long-term regret regarding the present consumption, even if today it seems like the right thing to do. (Look up hyperbolic discounting for more detail.)

2. Households typically exhibit various forms of mental accounting and are thus likely to handle the tax rebate differently than it would handle an equivalent amount of earned income, for example. Specifically, there is evidence that households are likely to consume more and save less of what is considered a windfall, and the windfall is likely to be spent on more hedonic and frivolous items. In this way, the government is encouraging consumption of frivolous items- just what we need in this country.

3. Households have almost all retailers on the planet screaming "Spend your rebate here!" in their faces. It's unclear whether advertising aids in utlity-maximization or just clouds the process. I myself just bought a food dehydrator from an informercial, so who knows.

4. The households for whom the rebate is most attractive are the ones who are most liquidity constrained. The rebate at a low effective interest rate might be a boon to them, but it could also be the case that the liquidity constraint was an effective self-control device.

Update: Whatever the reason, consumers aren't as forward-looking as Ricardo's theory would imply. From Greg Mankiw's blog: "Total sales at U.S. retailers rose a full percentage point in May as many consumers had more spending cash in their wallets from government rebate checks."