Print Friendly and PDF

The Bear's Lair: The coming deficit spiral

January 20, 2014

The colossal cost of much needed projects like U.S. infrastructure repair highlight the necessity to balance the budget.

 Policymakers and investors have been remarkably sanguine in the first few weeks of the year, appearing to believe that the problems bedeviling the global economy since 2008 are being alleviated and that the solutions produced for those problems will not themselves produce further and worse pathologies. Yet one area that confidence seems ill-conceived is that of global-fiscal balances. Governments are already planning new spending boosts for infrastructure (in many cases much needed) without considering that the current reduced budget deficits are a function of artificially depressed interest rates and will disappear when rates revert to a more normal level.

Not that the current budget outlook is any too bright, even with interest rates at current levels. According to the Congressional Budget Office's latest projections, which date from last May although they were confirmed by a debt management document in December, the budget deficit will total $560 billion in the year to September 2014 before falling to $378 billion the following year and ticking back up to $432 billion the year after that. However, you have to remember that this projection assumes full operation of the "sequester," which has already been negotiated away, in return for budget increases and spending cuts that fall almost entirely after 2020, i.e., too late to affect current incumbent politicians. 

Even in this projection—assuming the U.S. economy never runs into a recession—the deficit starts to rise inexorably after 2015, reaching $733 billion in 2020. As has been extensively discussed, the main sources of this rise are the upward trend in Medicare/Medicaid and Social Security spending. While healthcare is the bigger long-term problem, in the medium term the swing in the Social Security balance is especially damaging: the "off-budget" Social Security fund balance goes from flat in 2018 to a $100 billion annual deficit in 2023.

You have to remember that, as of January 2014, the time horizon of a Senator is at most four years, ten months until November 2018, the latest rollover date of the three current classes of Senators. The horizon of the President is three years until his term of office ends and that of House of Representatives members is a mere 10 months. Public Choice Theory, which looks at the real-world incentives applicable to the political class, shows that, constrained by such short time horizons, politicians' incentives are very different to those an ideal policy would require.

Unless there is mass popular demand for budget balancing or a major financial crisis that prevents borrowing, there is really no incentive for politicians to close budget deficits. In 2008, the financial crisis was used as an excuse to widen deficits further, financing them through the Federal Reserve’s money printing. Since no disaster has yet eventuated from this approach, politicians and the Fed have spent the last year congratulating each other on their successful suspension of the laws of economics.

Quite apart from the question of bursting bubbles, this isn't going to work for long, because interest rates are going up. The CBO has estimated that net interest costs in 2020 will be $644 million, just a little below the deficit that year. On the Brazilian definition of "primary budget balance," therefore, the U.S. would be just marginally in deficit in that year. However, in reality that projection looks excessively optimistic.

According to blogger Mike Shedlock's "Mish’s Global Economic Trend Analysis," even at an average 3% interest rate, interest payments in 2020 will total $668 billion. But if the interest rate averages 5% (a much more realistic figure, given inflation at the 2-3% level), interest payments in 2020 will total $1,285 billion. That would make the deficit that year reach $1,374 trillion—plus the effect of interest on interest from the higher deficits in the trajectory to get there. That would put the deficit in 2020 at 6.0%-plus of GDP in, we are to suppose, no less than the eleventh year of a by-then robust economic recovery.

It's not as if anywhere else is much better off. Britain's budget deficit is much worse than that of the U.S., and the country also has a Sword of Damocles hanging over it of baby-boomers retiring and a potential rise in interest rates. France's deficit in 2013 is projected to be the same as that of the U.S., according to the Economist. And France has adopted suicidal economic policies, including a 75% top tax rate that make its growth prospects much worse. Germany is running a tiny surplus, true, but it has a potentially huge contingent liability of the deficits in southern Europe if it wants to keep the Euro together. 

As for Japan, it has both a huge deficit and a catastrophically high level of public debt, which is why it has the most suicidal over-expansionist monetary policies of any country. Needless to say, a return to normal interest rates would definitively push Japan over the fiscal edge. Only Brazil may have interest rates higher than it needs, But its central bank just pushed them up further to a level close to 5% in real terms, while its government continues to spend on infrastructure for the 2014 World Cup and the 2016 Olympics, presumably under the impression that expensive sporting spectacles will at least take its citizens' minds off impending fiscal collapse.

Siren voices in all these countries wishing to undertake more fiscal expansion have pointed out that the countries' infrastructure is decaying (or in the case of Brazil, inadequate in the first place) and therefore a program of infrastructure investment, undertaken at a time of exceptionally low real interest rates, would both pay for itself and put the unemployed back to work.

There are two objections to this. One is budgetary, to which I shall return. The other is that in the U.S. infrastructure now comes at an appalling cost. Take, for example, the marvel of U.S. infrastructure investment, the interstate highway system. At its inception in 1956, it was projected to take 12 years to build and cost $112 billion in 2014 dollars. In actuality, it took 35 years to build and cost $510 billion in 2014 dollars—a cost overrun of over 300%.

However, let's ask the trillion-dollar question: do we think it could be built today for a cost of $510 billion? When you look at the California High-Speed Rail project, expected to cost $91 billion in today's dollars for only 800 miles of track, the answer is clearly no. Let's make the very generous assumption that interstate highways can be constructed for one-third the cost per mile of high-speed rail (don't forget that many such highways run through urban areas, as will the California High-Speed Rail, and that an interstate highway takes up more area and uses more resources than a rail line). Then to construct 47,182 miles of interstate highways at one-third the cost per mile of California's High Speed Rail would, if undertaken today, cost $1,794 billion. That's 252% more in real terms as the actual system cost in 1956-91 and 16 times as much as it was projected to cost. Oh, and I very much doubt if we could get it finished by 2100.

You can carry out this test with any equivalent infrastructure projects you like. The Holland Tunnel under the Hudson River was completed in 1927 for $48 million—$600 million in today's dollars—whereas a functionally identical, duplicate tunnel project was killed by New Jersey Governor Christie in 2010 because it was going to cost $8.7 billion. Like healthcare, but without the quality improvements, or like college education, the infrastructure sector has undergone an appalling decline in productivity in the last half-century. Given that, U.S. infrastructure projects, if costed honestly, almost never provide an adequate return on investment, whatever discount rate you use. Economically, we're right to let our infrastructure slowly decay, simply patching the worst holes. The cost of replacing it is unjustifiable.

Consultancy studies are urgently needed to precisely determine why infrastructure costs have behaved so badly–except that consultants themselves are very likely part of the problem and so would falsify the results (more Public Choice Theory.) Apart from consultants, one possible reason for cost escalation has been regulation, especially environmental regulation, which makes infrastructure projects take much longer to complete than was the case 50 years ago. British housing, subject to draconian planning requirements since 1947, has the same problem. The Financial Times has calculated that if chickens had risen in price as rapidly as British housing costs, prime broiler would today cost $85 a pound.

To build infrastructure, we must therefore subcontract it to the private sector, cut out its costs and use 1955 costings, adjusted for inflation, to prepare budgets. However, even in this case, we still should not add further infrastructure spending to our current budgets with their current deficits.

Public-sector accounting has many defects, not all of which are improved by aligning it with the private sector. It is really not helpful to announce that Medicare costs will increase U.S. debt to 600% of GDP by 2080. We know that won't happen because such a burden would be un-financeable. However, one private sector technique that the public sector should adopt is separating current from capital spending. Many advocates of infrastructure spending advocate this and exclaim indignantly at the short-termism that lumps the two. However, the corollary of separating the two is that the current spending budget should run a surplus sufficient to pay off the debt attached to it, with capital projects being financed separately and undertaken only if their revenues pay for their financing.

Thus if infrastructure spending is to be increased by 2% of GDP, as many have advocated, the balance in the remainder of the budget should be improved by fully 5.4% of GDP, including revenues from infrastructure, to pay for the infrastructure and eliminate the current deficit of 3.4% of GDP. Only when the non-infrastructure budget is balanced, including a sinking fund to repay the debt, and building costs have been reduced to their 1955 levels in real terms, can we afford to build new infrastructure.

In the meantime, I choose lousy infrastructure over national bankruptcy. Politics is full of these short-term choices. 

Gross Domestic Product (GDP) is a broad measure of the economy that measures the retail value of goods and services produced in a country.


Commentary Archive