Here's a short story, simple but true, that illustrates a lot of public finance: from public choice to Ricardian Equivalence.
I live on a cul de sac. Many of my neighbours wanted our gravel road paved, and were prepared to pay for it. The municipality held a neighbourhood referendum, and the motion passed by more than the required two thirds majority. They hired a contractor to pave the road, and split the cost equally among the 60 households. And today I got a letter asking whether I wanted to pay a lump sum of $6,050 or 15 annual payments of $583 added to my property taxes (which includes interest at 5%).
The first question is why did we need the government to do it? Why didn't we just hire the contractor ourselves?
I expect the answer is that the paved road is a non-excludable good. If one of the neighbours refused to pay, we couldn't really stop him using the paved road. Only the municipality could enforce payment. We couldn't prevent "free-riders". We might also add that the paved road is a non-rival good; it's a quiet street, with no congestion, so one extra car or household using the paved road has a marginal cost of zero, so it wouldn't be efficient to exclude anyone. Non-excludable, and non-rival; you could say it meets the classic definition of a "public good". But I share Frances' fear that the "public good" concept is a pedagogical bad (pdf). It's so easy to confuse "public good" with "publicly provided good".
A Coasian would argue that if the transactions costs were low enough, we could still have gotten the street paved without needing the municipality to intervene. If the sum of the neighbours' Willingness To Pay exceeded the cost of paving, we would have been willing to pay the costs. And if it didn't, the road shouldn't be paved anyway. But of course it's hard to get people to reveal their true WTP if what they actually pay depends on their announced WTP. I'm not up to speed on the latest findings in the design of revelation mechanisms which might solve this problem. But aren't they vulnerable to "hold-up" problems: where someone commits in advance to refusing to pay (if they can do so credibly)? Or arguing about which of several potential mechanisms will be used?
In any case, we didn't take that route.
But there was an earlier free-rider problem. Things don't just happen by themselves. Someone had to go around to organise the neighbours to lobby the municipality to hold a vote. Like toilet cleaning and department chairing, it's a job that many wanted done, but everyone would prefer someone else to do. Who would be the one to do it?
Simon lives near the end of the cul de sac; has two young boys who wanted to rollerblade; and likes to keep his car clean. It was therefore common knowledge that Simon was the one who most wanted the street paved. Simon was also unable to pre-commit to not doing the job of organising the neighbours, which would force someone else to do it. So Simon must have reasoned as follows: "if an equilibrium exists in which one person does the job, that person will be me; so if I want the job done, I will have to do it myself; and so I had better do it".
Or maybe he is just civic-minded.
The vote was nearly unanimous. That puzzled me at first. Surely some people wouldn't personally care whether the road was paved. Some might even prefer it unpaved. But then the paved road should increase the values of the houses on the road. Even if I personally didn't care whether or not I was living in a house on a paved or unpaved road, if it increased the value of my house by more than the cost of paving the road, I could still benefit by selling the house, pocketing the capital gain, and moving to the next street over. What matters for the house prices is not the value that current residents place on having a paved road; it's the value that the marginal potential resident would place on having a paved road. According to the Tiebout model, the personal preferences of the current residents wouldn't matter at all; population flows ensure that every municipality should provide the optimal mix of goods and taxation. There are transactions costs of trading houses and moving, of course. But some might have been considering moving anyway.
So now we get the bill. But we have a choice of how to pay it. We get to choose, as individuals, whether the government expenditure will be "tax-financed" (one lump-sum payment), or "bond-financed" (15 annual payments with the same present value as the lump-sum, to retire the debt plus interest).
Ricardian Equivalence says that the method of finance of government expenditure doesn't matter. You can see the point. If I have $6,050 cash, I could either pay the lump-sum, or I could lend it to the municipality (or lend it to the bank which in turn lends it to the municipality), pay $583 per year extra taxes to the municipality, which gives it straight back to me (perhaps via the bank) to pay down the loan. Or if i don't have the $6.050 cash, I could borrow it from the bank (or from the municipality, which in turn borrows it from the bank), then pay the bank (or the municipality, which in turn pays the bank) $583 per year to pay off the loan.
At first sight, it's a wash either way. Whether the municipality financed by taxes, borrowing, or gave us the choice between the two, shouldn't matter.
But small things may make it matter. Now we go from public finance to personal finance. I have until the end of May to decide. Here's what I'm thinking about my choice.
If I knew I weren't going to move house in the next 15 years, I would pay the lump sum. I have no debts, am maxed out on my TFSA and RSP, and can find the $6,050, so a tax-free and risk-free return of 5% looks pretty good to me.
But suppose I sold. Would my lump-sum be reflected in the price at which I could sell my house? It ought to be, if the housing market were efficient, but is it? I don't mean "efficient" on any macro-scale, but at a micro-micro level. Would my house in June with the $6,050 already paid be worth $6,050 more than my same house with 15 annual tax installments of $583 still due? I can think of a few reasons why it might not be.
Maybe (some) potential buyers ignore the property taxes. If so I should not pay the lump sum, if I might move soon.
Maybe (some) potential buyers look at the property taxes, but don't look at the fact that the extra taxes last for only 15 years. Then I should pay the lump sum.
Or maybe, (some) potential buyers look at the property taxes, then multiply by (say) 100 to get a rough estimate of the assessed value of the house (we have property taxes based on market value assessment here). Then I should not pay the lump sum.
Or I could just say "To hell with it; financial intermediation is always imperfect, and the right thing for me to do is to minimise the stress on the global financial system, so I should just pay the $6,050!"