Tuesday, July 28, 2009

Why Borrow When You Can Pay Now?

The decision to borrow money, especially when that decision is made by a government, must be made with great care. People's attitudes about debt are colored not only by cold calculations about costs and benefits, along with predictions about the course of an uncertain future, but by deeply ingrained moral attitudes about the very notion of being obligated to another person. "Neither a borrower nor a lender be" seems like such sage advice that people will go to great lengths to avoid taking out loans. Yet we know that people, businesses, and governments regularly borrow, in good times and bad.

If people and institutions did not continue to borrow money, there would have been no need to save the financial system from its recent (and ongoing) troubles, because people would not need financial institutions at all. They also -- it should be pointed out -- could not save money with interest, because that would require that someone else take temporary control of their money with the contingency that it be paid back with interest. A modern economy's very dynamism, in fact, depends crucially on the existence of markets for credit and debt; and admonitions not to be a borrower should best be understood either as archaic misunderstandings or as cautionary advice not to go too far.

Last week, I published two blog posts (here and here) extending my frequent discussion of the public's perverse and misinformed attitudes about federal deficit spending in the U.S. The discussion was divided into two pieces: first, the conditions under which it is wise for the federal government to borrow money during an economic downturn, and second, the conditions under which it is wise for the federal government to borrow money when the economy is healthy. In the course of the latter discussion, I claimed that it is always wise for the federal government to borrow money -- even during good times -- because there are always projects available to the federal government that will pay off in the long term in amounts greater than any debt (and interest) taken on to finance those projects: "[S]ome things are so valuable that it makes sense to borrow money to buy them."

Summarizing that argument, I made the following assertion: "[T]here will always be government projects available with rates of return that exceed borrowing costs." I made that claim not to cover the universe of theoretical possibilities but to describe the world in which we should expect to live for as long as the current economic system continues in this country. While it is possible to describe a situation where the government has exhausted all of its high-value investment opportunities, such that no further borrowing would be wise, that is not going to happen. (I wish it would. If it did, then of course our policy choices should change.)

Several commenters on my second post from last week also brought up the question encapsulated in the title of today's post: Sure, we can borrow to finance projects with long-term rates of return that exceed borrowing costs, be we do not have to do so. We might want to pay up front for these projects. Should we?

One possible reason to run balanced budgets during good times is as a show of good faith and to demonstrate our ability to discipline ourselves, essentially to reassure borrowers that we are good credit risks. One response to this is that the U.S. federal government has never defaulted on its debt, and its debt securities are treated as the equivalent of cash by shrewd money managers the world over, even in the face of a crisis that has shaken the global economy's foundations. Even so, with rumblings that foreign lenders might be losing confidence in our long-term ability to repay debts, maybe time is running out.

If that turns out to be true, however, it will not be because we are taking the advice to engage in borrowing to finance long-term investments but because we have allowed the health care system to spiral further out of control, destroying both government and private finances in the process. More to the point, it is difficult to see how any potential lender would be reassured by a government that decided not to engage in long-term investments in the name of fiscal responsibility. A reasonable lender would take that as evidence that the U.S. government is run by people who do not understand basic finance.

This still, however, leaves the possibility of engaging in those long-term investments but paying for them with tax dollars rather than with further borrowing. What are the implications of such a decision? The government would be taking tax dollars from people today, using those dollars to buy something today that has benefits mostly (or completely) in the future. If the government had not collected those tax dollars, the people or businesses from whom the taxes would have been collected could have used the money to consume or invest in the way that they found most appropriate according to their circumstances. By paying taxes now, they would lose those opportunities in the name of not adding to the debt. It is important to remember, however, that the long-term debt burden would go down even if the projects had been financed by borrowing.

In other words, the decision to borrow or not to borrow boils down to the present-versus-future dynamic that we hear so much about, but in reverse. Forcing ourselves to pay for investments up front means that future citizens will benefit not just by receiving the benefits of the projects in which we invest but by receiving those benefits at the expense of our current sacrifices. Reasonable people can differ on how much we owe future generations, but it is at the very least not obvious why we are morally required to take that extra step.

In addition, if we force ourselves only to engage in those long-term investments that we can currently finance, we will surely end up financing many fewer such investments. I would prefer not to tell our grandchildren that we passed up profitable investments because we could not pay for them up front, when financing would have been available. Their patience with our explanations that we were just being prudent will surely be limited. As well it should.

-- Posted by Neil H. Buchanan


Michael C. Dorf said...

We can agree with all of this (and I do) but still have the question I raised on the last post: if the future benefits of a cost-justified loan are not monetizable, then the borrower (whether private or public) cannot service the debt from revenue produced by the project. Unless I misunderstood this post (always a possibility) you don't address this liquidity limit on borrowing.

Neil H. Buchanan said...

Mike's question can be interpreted in two ways, I think. One possibility is to say that he is simply defining an empty set. That is, if the measure of fiscal rectitude is a declining debt-to-GDP ratio, then the benefits of an investment have to show up in GDP for the spending project to be classified as public investment.

This is not quite right, however, because GDP currently includes some things that are not "real" in the same way that Mike suggests. For example, the value of owner-occupied housing is included, even though there is no transaction and thus no money changing hands.

Therefore, the assumption must be that the measure of GDP includes at least enough things in it that can generate tax revenue to keep the path of debt/GDP falling. If GDP includes some things that cannot generate tax revenue, then the remainder must be sufficient to make up the difference.

My post today very indirectly addressed this, believe it or not. When I said that I am trying to deal with real-world possibilities and not extreme theoretical cases, I meant to include a variant on the type of hypo that Mike poses. That is, I am willing to assume that we will never be in a situation where the only investments that we make are illiquid and where the overall tax system cannot generate enough money to cover the additional debt. (Note that, if we do make investments that have non-monetary payoffs, then by assumption we must still believe that they are worth paying for, else we wouldn't think of those payoffs as benefits.)

Jonathan Noble said...

Has anyone ever done a history of the U.S. public debt, why some administrations increase the debt, why others commit themselves to shrinking it?

Michael C. Dorf said...

Hmm. I read Neil's response as a way of saying that most things really are monetizable. Take health care. Enabling people to live longer lives may actually be costly, because the extra years tend to be non-productive. So if the govt borrows $100 billion today to generate technology that enables people to live 2 years longer (on average) but does not extend their productive lifetimes, by my reckoning, the $100 billion plus interest must be paid back from other revenue. I read Neil to say that if the extra two years really are worth it, then the beneficiaries will be happy to pay for it through higher taxes later. But that's my point: There are only so many such projects that can be funded from other sources before those other sources themselves run out. I read Neil's assumption at the end of his response to my first comment to concede this point but to add that in a large enough economy, this will not likely be a limiting factor.

egarber said...

So if the govt borrows $100 billion today to generate technology that enables people to live 2 years longer (on average) but does not extend their productive lifetimes, by my reckoning, the $100 billion plus interest must be paid back from other revenue.

But in all likelihood, those investments also would lead to higher productivity during their active years. Plus, estates and investments would continue to grow for those folks during their extended life span. I guess you could counter that by saying they would potentially be a drain on social services if they failed to effectively manage their finances. But in the bigger sense, I think even in your example it's possible that the investment would pay for itself.

Neil H. Buchanan said...

I think Mike's point is partly answered by egarber. In addition, it is worth noting that any problems along these lines would show up in the cost-benefit computation that determines whether the project is an investment or not. Admittedly, that process would be imperfect; but the concept of this type of net present value computation is precisely to separate those projects that "more than pay for themselves in the long run" from those that do not.

This actually is an important additional point for my longer-term project, because Mike's observation (and egarber's answer, along with my answers) suggest that some things that do not pass the cost-benefit test would still be valuable in a more profound sense. Thanks!

Neil H. Buchanan said...

In response to Jonathan Noble's question on this post and his comment on last week's post:

-- I am not aware of any studies that specifically ask "why some administrations increase the debt, why others commit themselves to shrinking it." Could be a good independent study project when you get to law school!

-- After suggesting that paying off deficits serves a signaling function, Jonathan added: "What else could be the impetus behind the push during the Clinton years to erase the deficit?" I think the better explanation is that Clinton simply capitulated to the orthodoxy on Wall Street (Robert Rubin), mostly to try to limit the political damage that he perceived from the Perot phenomenon and the Contract on America. I don't think it had anything to do with a desire to prove continued credit worthiness -- or with any non-political principle of any kind.

egarber said...

suggest that some things that do not pass the cost-benefit test would still be valuable in a more profound sense.

Right. I mean, it would be tough to do an empirical cost-benefit analysis on say, a project to raise awareness about racial inequality. But at the same time, it's hard to argue that such an effort doesn't generate value.

I suppose maybe there's another question: does the flavor of benefit dictate whether we should borrow for it -- vs. paying in real time? Probably not, but I'll let Neil tell me whether it's a valid question :)

egarber said...

And I finally added a friggin' profile pic :)