Phantoms in the Dark Part 4

I’ve been a bit under the weather for the last few days, but I didn’t abandon the “Phantoms in the Dark” project.  I did spend a completely unreasonable number of hours squinting at my computer screen, slogging down words on the Abba Lerner and James Buchanan alternatives.  One says that federal debt can’t hurt future populations, and the other says it must hurt future populations; I think both are kind of true, which is possible only because in spite of using similar words they are looking at different things.

And then I threw it all away, because in the end the basic arguments are pretty simple.  They are also in some ways a bit arcane and even to some extent irrelevant, or at least they seem so to me.  I’ll say why after I describe them. 

As I said a few posts ago, the Lerner faction is looking at allocation of resources, and the Buchanan faction---or at least Buchanan himself---is looking at human satisfaction.   He’s looking at people’s position on their utility surfaces (if that phrase is foreign to any reader, just ignore it and stick with “human satisfaction”.)

Both sides, or all sides, of the discussion on this would like to isolate just the core issue, which is the question of whether using debt rather than taxation to pay for something really moves the “burden” of payment through time.  So both sides would like to set aside any other effects this decision might have, such as an impact on inflation or unemployment or economic growth.   This is a little strange, but ok, we’ll (temporarily) vaporize those things up front by just assuming them away.

The essence of the Lerner-side allocation-of-resources argument is that we can’t reach forward in time, steal the resources that will be available in 50 years and use them for government purposes now.  Or for private purposes either, for that matter, or I’m sure some enterprising entrepreneur would have offered that service long ago.  Those future resources are sealed behind the impenetrable barrier of time, and we can’t get at them.  If the government is going to do something it has to use resources that are available at the time the government activity is undertaken.  If, for example, the government decides to build a ship it must hire people, buy steel and tools and fuel and all the other inputs it needs to complete that task.  Those people could be doing other things.  They could be making jewelry, or dealing blackjack at the nearby casinos, or teaching economics at the local university.  If the government is going to build the ship now, it needs to hire the labor that is available now, and the nation will have to forego any alternative current uses of that labor.  In that sense the material burden of building the ship in terms of diverted resources falls on the time the ship is built, not on some future time when the actual financial payment will be made.  The payment has an impact, but (since we’ve vaporized the notion of unemployment and inflation and economic growth) that effect is just a redistribution of cash toward those who hold the bonds when the nation chooses to pay them off.  Those people may use the money to increase their consumption, and those who are taxed may choose to decrease their consumption, but losses match the gains exactly, so the nation as a whole is no better or worse off than it was before.

I think this basic argument is inarguably true within its artificial world of limiting assumptions. 

But hold on, the Buchanan personal-satisfaction side says; you can’t just look at the distribution of physical stuff to figure out where a burden lies.  Burden should not be defined like that.   When a baker trades some loaves of bread to the tailor in exchange for some clothing, physical resources have been shifted around, but everyone is actually better off.  The tailor can eat, and the baker can get dressed.  And the personal-satisfaction side will point out that in the current period, when the government offers bonds, no one is forced to buy them.  Those that do buy them and give up cash that they could use for other purposes must be better off after that transaction, not worse off, or they would not voluntarily do it.  They bear no “burden” in personal-satisfaction terms.  And when the government buys the things it needs, labor and physical inputs, to build its ship, those transactions are all voluntary too.  Everyone is more satisfied in the current period than they were before, so the current period bears no satisfaction-burden anywhere.  But in the future, if the government pays off the bonds by taxing the general population, that transaction is not voluntary, and imposes a direct satisfaction-cost on the taxpayers.

And this, too, seems to me to be inarguably true within its limiting world of assumptions.

The problem with both of them, to me, is that I think the limiting worlds they describe are impossible, that the limiting assumptions can’t all hold at the same time.  But these economists were or are all very smart people.  I think they were, and are, aware of this.  Here’ how Buchanan put it here:

“Fiscal theory must always recognize the fundamental two-sidedness of the government's fiscal account. It is not methodologically permissible to examine, for example, a change in the level of taxes without examining, at the same time, the offsetting or compensating changes on the expenditure side, provided that the quantity of money is held unchanged… It would do the engineer little good to attempt to analyze the movements of one side of the teeter-totter on the assumption that the other side remains unchanged, for the other side must change to allow for any initial movement.”

Yes.  But when we do that, the whole issue changes.

When we say that the government has chosen to build a ship (presumably for defense), or or to build highways or sewage treatment plants or levees to hold back flood waters, these things take years to build and once built will last for decades. The population of the period during which the expenditures are made do not get a direct benefit from those expenditures; all of the benefit is felt by future populations.  So why should those future populations not be taxed to pay for them?  It could be argued that what we should do in cases like that is borrow the money now, and raise taxes enough to make payments over the life of the asset we build to pay that loan off.  In effect, that we should take out a mortgage to be paid off over the lifetime of the flow of general benefits from the expense.  But what is the lifetime of benefits for investment in the internet, or in research into affordable solar power, or a the discovery of a vaccine for polio?

And of course, the initial assumption that nothing we are doing will alter the trajectory of economic growth is utterly wrong from the start; damned near everything we do alters the trajectory of economic growth in one way or another.  Every time we have a new idea, or change our buying behavior by switching to bulk buying at Costco, or flock to Twitter to pass notes to the universe, that changes the future, often in ways we can’t begin to predict.  So once again, if something the government does now increases growth, or the potential for growth, is it a net burden to future populations who benefit from that growth to borrow the money to do it? 

Also: when we talk about future “generations” we seem to be talking about time periods fairly far apart from each other.   But the distant future is hard to predict; when we get to that future it may be hard for us hobbling creakers to even comprehend.   Even a period as short as 25 years is hard.  25 years ago was 1987: who, in that year, would have predicted the dot-com bubble?  No one even knew what a dot-com was; Timothy Berners-Lee and Robert Cailliau did not publish the proposal to create a hypertext presence on the internet until late in 1990, and the first publicly available WWW presence on the internet was established on August 6, 1991.  And who, in 1987, could have expected to survive long enough to be able to play a game on a telephone, much less a game like Angry Birds or Fruit Ninja on a wireless telephone they carry in their pocket?   And if small things can change the trajectory of the future as profoundly as that, how much more can we change the future by actually investing today, even if we have to borrow to do it, in new research (the Web was enabled by the existence of the internet, which emerged from government research in the 1960s and 1070s).   Laws, infrastructure, social expectations, technologies, methods and views about the distribution of incomes, can all change between now and the future time we are considering, and fine calculations of positions on utility surfaces that we make now may be simply swamped by other changes that are created by current actions, even if the actions seem small now.  It may be that the best we can do is to just make sure that nothing we are doing now will damage the production potential of that future time, or reduce the sheer quantity of consumption-plus-investment that is available in a future we don’t and can’t comprehend.  

And of course, employment is not always full---in fact it’s never completely full, or we would have runaway inflation.  But there are times, such as now, when employment is very far from full, and that is imposing enormous costs on the current population in areas of high unemployment, and also on future populations who will not be able to afford to go to college, or who will have to endure childhoods spent in greater poverty than they otherwise would.  How these deprivations cascade down through time is a question far larger than I can answer, but there is no doubt that they do.

And finally, at least for today, much of the discussion of debt takes it as given that the alternative is taxation.  But the government has the power to create money.  Why not finance current government expenditures that way, rather than borrowing or taxing?  Buchanan recognizes that possibility too, particularly during recessions.  He says, in Chapter 9 of the online book I linked to above, in a situation where there are idle resources,

“The actual private disposition over current resources need not be reduced. To provide the monetary means for its actual purchases of resource services, the government may create money quite readily. The efficient means of purchasing the services of unemployed resources is through inflation of the currency. Individuals owning the resources concerned are made better off, and no one in the economy is made worse off.”

And the emphasis is his, not mine.  Just to be clear, when he says currency inflation he means making money out of thin air, what is often popularly called “printing money”.  Lerner too---emphatically---recognized the possibility of using newly created money to finance government purchases, and I’ll talk about that, the long promised functional finance question, tomorrow, trying to figure out something about the limits of government debt, rather than just who bears its burden. 

The problem with Buchanan’s prescription about what to do during a recession, though, is that it is politically impossible, unfortunately. How many of his political allies in the fight against public debt would agree with his prescription for simply creating money during a recession to finance the use of idle resources?  How many would even understand it?