Models of Fiscal Policy, the Trade Deficit, and the Dollar

Paul Krugman writes:

Robert Rubin is wrong about the dollar: He says: “You could have had [budget] surpluses that affected the savings rate and would have helped the trade balance. I think you would have had more confidence in the policy framework and you would have had a [stronger] dollar.”...

This is what John Williamson of the Institute for International Economics calls “the doctrine of immaculate transfer.”... Here’s a (somewhat) plainer English version:

The problem becomes apparent if one asks how a higher savings rate translates into a smaller trade deficit. It is not enough to insist that the accounting ensures that it must. A consumer deciding between a Ford and a Honda cares nothing about the US’s national income accounts. How does a lower US budget deficit persuade Americans to buy fewer foreign goods and foreigners to buy more US products?...

The chain of events would look something like this: a fall in the budget deficit reduces demand in the US economy; to avoid a recession, the Federal Reserve lowers interest rates; as a result, the dollar falls; this lower dollar makes US goods cheaper compared with foreign substitutes, causing the necessary switch in expenditure.... [I]t is naive to imagine that changes in the government’s financial balance can translate directly into changes in physical trade flows, without working through a mechanism such as the exchange rate. That is the fallacy of ‘immaculate transfer’ - confusing the accounting principle which says that the current account balance equals the savings-investment balance with the process that enforces that constraint on decision-makers...

And apparently the old fallacy continues to hold sway.

I think that Rubin is implicitly working in a different model than the NIPA-based monetarist workhorse that Krugman (and I!) instinctively reach for first. I think that in Rubin's mind the chain of causation looks something like this:

1. A government establishes a sane, balanced long-run fiscal policy.
2. Businesses conclude that the country is a safe one in which to invest to build export capacity: since they do not fear future random and confiscatory taxation or inflation, factories making internationally-traded goods that would have been built abroad are built here at home instead.
3. With more export capacity at home and less abroad, exports rise and imports shrink at the current exchange rate.
4. Supply and demand leads the domestic currency to appreciate in order to balance trade.

Rubin's channel is: good fiscal policy --> expanded export supply --> export surplus --> higher currency value.

By contrast, Krugman's channel is: good fiscal policy --> lower domestic interest rates --> reduced currency value --> export surplus.

Which side am I on? I tell my undergraduates:

• At a time horizon of 0-3 years, be a Keynesian: the most important things are the fluctuations in unemployment, in real demand, and in capacity utilization.

• At a time horizon of 3-8 years, be a demand-side monetarist: you can assume (provisionally) that fluctuations in employment, real demand, and capacity utilization die out; the most important things are the fluctuations in the composition of real demand (investment vs. consumption vs. government vs. net exports) and in inflation- and deflation-causing nominal demand assuming (provisionally) stable growth of the economy's productive capacity.

• At a time horizon of 8 years or greater, be a sane supply-sider: the most important things are the processes of investment in physical, human, and organizational capital that raise the economy's productive capacity.

Thus I was happy telling my undergraduates in 1985 that the reason the dollar was strong was because of the five years of Reagan deficits--high domestic interest rates, you see, pushing up the value of the dollar (and raising the trade deficit). And I was happy in 1992 telling my undergraduates that the reason the dollar was weak was because of the twelve years of Reagan-Bush deficits--large budget deficits starving the economy of capital that made us less productive than in some counterfactual in which we had elected some Eisenhower Republican in 1981.

And so today I call this one for Paul Krugman: we are only in year six of the Bush II derangement of American fiscal policy, and so I think the dollar is a little higher than it would be if the U.S. budget deficit were lower. But in two years I may have a different answer.

Comments

[Deepish Thinker]

The three time frame system is a fascinating approach and may have considerable utility when it comes to passive analysis, but the implications for policy advice are somewhat disturbing.

The clear implication is that the first question the president's economic adviser should ask is, "What is your time frame?" The answer would then determine whether Keynesian, demand side or supply side policy prescriptions are offered.

Thanks to the election cycle, the president's time horizon is almost always going to be in the 0-3 year range, which suggests a strong bias toward Keynesian advice and a systematic neglect of the supply side issues that actually matter most in the long term.

This may actually be how economic policy is formed, but the thought is somewhat depressing.

[Bernard Yomtov]

Brad,

You should be more careful in your terminology. Soon we will be seeing everywhere that "even Brad DeLong supports supply-side economics."

[Barkley Rosser]

Well, of course in the shorter run, exchange rates are driven by all kinds of wacky noise, bubbles, and third rate random walks, with indeed a longer term convergence to PPP equilibrium, which would support your argument about the longer term, which does indeed seem to be in the data (look at relative exchange rate movements over a half century).

BTW, a three period framework shows up in certain literature on financial markets. Thus, in the very short run, look at econophysics kinds of technical patterns related to peculiarities in short run frequencies (espcecially within a day, and not more than a week). In the medium term, months, even maybe over a year, think behavioral finance and anomalies and goofy stuff. In the longer term, go for fundamentals and textbook stuff.

[Deepish Thinker]

Bernard has half a point. Were the good people at Fox News to get hold of this post they might well interpret it as further support for their monomaniacal obsession with tax cuts.

However, regardless of how you feel about Fox News, it is neither fair nor accurate to "supply side" as a pejorative term. Whatever the connotations the term "supply side" carries in the political arena, it means something quite different to economists.

I doubt there is an economist anywhere who would seriously argue that, "the processes of investment in physical, human, and organizational capital that raise the economy's productive capacity", aren't important to long term economic outcomes.

[Bernard Yomtov]

DT,

Of course "the processes of investment in physical, human, and organizational capital that raise the economy's productive capacity" are important to long-term outcomes. Who imagines otherwise?

But I think you, and Brad, should recognize that, whatever meaning economists assign to the term "supply-side" has been overrun by its political meaning. There are few people who, upon reading that X "favors supply-side policies" will conclude other than that X favors endless tax cuts and believes they are, at a minimum, self-financing. Regrettable, but true.

[christofay]

Well, who cares about Rubin? What was he doing in the intermediate term while Citi was extracting possible long term profits in an immediate Enron manner?

Can he please borrow a billion xera so he can purchase a traditional super senior security in Citi-another Sheik of Arabia special-but before the collapse rather than after?

The more bailing in he can do the less bailing out that we will have to pay for.

His mistake was taking a position on a landmine while the Greenspan-Bush era was still underway.

[Robert Waldmann]

Weld minds think alike. I had a vaguely similar defense of Rubin. I didn't get the FDI angle.

Note that a one off burst of FDI should affect the balance of trade but not the current account as foreign investors repatriate profits. Your version of Rubin's argument requires a shift in fiscal stance to cause a constant flow of FDI (to make the strong dollar consistent with floating that is a balanced balance of payments).

This is possible but it tends to undermine your trichotomy. If a shift can cause a flow, short run events can change long term trends.

[Charles]

Gee, and I had just written an explanation for why *Krugman* was wrong (tinyurl.com/2ot8qw).

Basic gist of the comment: Krugman mistakenly believes we are always near equilibrium. In reality, radical adjustment of the dollar might easily lead to a non-linear effects like the breakdown in the Pax Americana, catastrophic loss of human capital, and large-scale waste of resources. Furthermore, unlike Argentina, a crash of our economy would have profound impact on our trading partners. Finally, the likelihood of the emergence of destructive competition between nations along the lines of what occurred in the Great Depression (e.g., Smoot-Hawley), is likely.

Because the earth itself is in danger from global warming, loss of biodiversity, etc., the precautionary principle forbids us from doing the experiment.

[Measure for Measure]

Great 3-part summary of macro considerations.

"At a time horizon of 8 years or greater, be a sane supply-sider: the most important things are the processes of investment in physical, human, and organizational capital that raise the economy's productive capacity."

Technology may merit a mention here. (Yes, this might be implicit, but we should try to get the meme right.)

[wml]

Willing to flesh out the evidence behind those time horizons?

[piotr]

"in reality, radical adjustment of the dollar..."

Hm, this is the larger swing down in the exchange rates that 20 years ago, if not at any time after the gold standard. So what calamities are upon us?

Breakdown of Pax Americana may be looming, but that is because of Bella Americana policies. Other catastrophes seem a bit far-fetched. Far-east money did not exactly abandon USA, although it may change.

How about this theory on fiscal deficit: if our currency is kept artificially high in respect to Far-East currencies, this is done because of the intervention of the central banks in East Asia, and these banks have to park a huge amount of dollars somewhere, and this would have a big preference for government securities.

If we were running fiscal surpluss, it would be much harder for foreign central bank to keep dollars. They would actually had to expose themselves to the vagaries of our market. Remember how Japanese were investing in American real estate in 1980s and lost a huge bundle? (I guess that domestically they did even worse, so the diversifications was not all that bad).

If this is a sound theory, then it is helpful if the government saves when the people do not. Foreign credit will then go to individuals and companies and thus it will be exposed to larger risks, and this would dampen the oversupply of foreign credits, and, by extension, the merchandise deficit.

With lesser supply of foreign credits we could have somewhat higher interest rates, better incentives to save (as individuals) and less incentives to increase debt.

Dollar would have smaller tendency to get overvalued and investing in production capacity in USA would be more competitive.

To recap, if the behavior of foreign central banks is not driven by conventional profit motif and it affects the trade balance to such a degree that the economy experiences a big shift between different sectors, then the choice of policy has to take it into account.

In particular, budget deficit could decrease interest rates and stimulate manufacturing outside our borders, or more precisely, make it easier to intervene in financial markets to achieve it.

[Charles]

Piotr, there are parts I agree with and parts not. The key government functions are providing defense, infrastructure, and a framework of social fairness, such as a justice system, universal education, and a pension system. Well-executed, these allow the economy to perform at its top potential and therefore generate the highest GDP and the lowest tax burden.

But suppose that the income so generated went mostly to foreigners, leaving Americans poor. This would be a politically unstable system. So, another function of government is to ensure that its citizens receive a large enough fraction of GDP + income from external investment to keep them happy.

This is where exchange rates become important. If the dollar falls by 90%, not only will Americans have to cut back on buying oil and Chinese toys, they won't be able to afford the basics. Food, instead of being sold internally, will be exported. Kids will stop going to school and will work. Infrastructure will decay.

Within some broad limits, exchange rates can vary without causing major damage. But in fact we have faced a series of catastrophes due to the decline of American living standards post-Reagan. Our heavy manufacturing base is gone, and with it, the knowledge to do manufacturing. Our infrastructure is in poor shape, well behind most of the industrialized world. And most critically, we have begun to lose out democracy.

Maybe more than begun.

[Majorajam]

Question: at what horizon do the multitudinous trillions in involuntarily held dollars matter? Much of this has been stockpiled just in the last five years and, well, it's a lot of eggs in one basket for a lot of not so wealthy countries. Their faith in the productive capacity of the United States and vitality of its consumers is presumably important lest they influence the voting machine by 'diversifying' at every opportunity, or worse, stampede for the exit at the first whiff of panic- (perhaps a worthy counterfactual could be an administration that stresses the importance of sound lending and reserve policies to its regulatory agencies, rather than stresses their subservience to the industries they ostensibly regulate- see in particular the OCC/SEC & Fed). Relatedly, any discussion of the effect on the dollar of policy or of the currency full stop that ignores its reserve status is lacking for sure- exorbitant privilege, dark matter and all that. I guess all that is my way of saying I tend to find practitioners' arguments more compelling in this arena than economists'.