November 23, 2015

Would a MAC Raise Interest Rates, Hurt Consumer Demand, and Slow the Economy?


Largely because of an overvalued dollar, America has suffered a virtually unbroken string of trade deficits for nearly forty years. The dollar will return to an equilibrium rate that balances imports and exports only when an automatic link has been re-established between exchange rates and balanced trade in goods and services.

Starting in the 1970s, the link between the dollar’s value and the exchange rate needed to balance trade was gradually destroyed as global trade in capital assets overwhelmed global trade in real goods and services. Today, as the Wall Street Journal recently noted, “Currency values are largely determined by central banks and capital flows,”[1]   For example, enough foreign exchange is traded across America’s borders in half a day to finance our trade deficits for an entire year, and enough comes in during less than two weeks to finance all American imports and exports for an entire year. The remaining fifty weeks are dedicated, shall we say, to other purposes.

Because of this fundamental shift in the way exchange rates are determined, any connection between the market exchange rate established by financial account transactions and the exchange rate needed to balance America’s external trade is nothing more than a happy accident – one that has not happened for nearly forty years!

To help fix this growing problem, I have proposed a Market Access Charge (MAC) that would automatically moderate the flow of capital coming into the United States whenever flows reached the point that excessive foreign demand for dollars and dollar-based assets had pushed the dollar so high that America was running trade deficits of one percent of GDP or more.[2]

While agreeing that steps such as a MAC should be taken to restore America’s international competitiveness with a fairly valued dollar, some have expressed concern that reducing foreign capital inflows would tighten the domestic credit supply, causing interest rates to rise, and reducing consumer demand, especially for large-ticket items such as automobiles.

Based on historical data for the U.S. motor vehicle industry, this note demonstrates that the MAC is highly unlikely to hurt consumer demand and economic growth in this way. In fact, implementing the MAC would greatly increase both domestic and foreign demand for made-in-America automobiles and other goods.

Automobiles – A Highly Relevant Case Study

The relationship between interest rates and the demand for automobiles is especially relevant to this discussion. In theory, the demand for consumer durables such as cars should be highly sensitive to interest rates. According to the National Automobile Dealers Association, “Because the vast majority of new vehicles are financed, credit is crucial to the survival of the industry.” [3]    Furthermore, the total amounts paid and financed when buying a car are generally quite large relative to normal household purchases.

The automotive industry is also highly important to the U.S. economy, generating more U.S. merchandise exports than any other sector including petroleum, aerospace, chemicals and semiconductors;[4] one third of the value of all durable manufactured goods; and ten percent of all employment in durable goods production. Including upstream and downstream jobs, the auto industry provides jobs for about 8 million people who, on average, receive about $62,000 per year. [5]

If it can be shown that the automobile industry would benefit from a MAC – even if interest rates should rise –  we can be quite certain that the overall impact of a MAC on the American manufacturing sector would also be positive.

A Paradox: Rising Interest Rates and Rising Auto Sales

Contrary to what might be expected, auto sales have tended to rise when interest rates are rising and to fall when rates are falling. Furthermore, this correlation is relatively high and consistent over time.
(Figures 1 and 2). Are car buyers crazy?

Figure. 1

Figure 2
No. As is well known, a statistical correlation between two variables does not necessarily mean that one of the variables is driving the other. Car buyers clearly do not buy cars just because interest rates go up. Nor do interest rates go up just because of a rising demand for cars.

Paradox Resolved: Consumer Sentiment Rules

 Car sales tend to rise along with rising interest rates for the following reasons. First, car sales are driven by consumer sentiment – the degree to which consumers are optimistic about the future and thus about their ability to repay car loans. Second, consumers are most likely to be optimistic if the economy is growing and unemployment is falling. Third, when the economy is growing, demand for final consumption goods and for investment goods will be rising. This in turn increases the demand for capital to finance both consumption and investment expenditures. Consequently, both prices and interest rates tend to rise together – at least until the Fed steps in.

Consumers know that the Federal Reserve, at some point, is very likely to raise interest rates to prevent the economy from overheating. Consequently, to avoid paying even higher interest rates on new car loans, consumers quite rationally buy today rather than waiting until tomorrow. Hence the positive correlation between higher interest rates and higher new car sales.

This relationship between consumer sentiment and rising car sales shows up clearly in Figure 3. Although the correlation here is not as tight as it was between interest rates and car sales, the relationship at least works in the expected direction – when people are happier and more confident, they buy more cars.

Figure 3
The relationship becomes much closer when we look at a shorter and more homogeneous time period – say from 2000 to the present (Fig. 4). This chart is essentially the same as Fig. 3, but the values have been connected so that the sequence from one year to the next is easier to see.

Consumer confidence started out rather high at the beginning of the century when the glories and excesses of the Dot-Com Bubble still dominated public thinking. However, as the realities of the bursting tech bubble sank in and people turned their attention to flipping houses, car sales began to drop.

Then came the Crash of 2008. The bottom dropped out of the car market, which plunged from over seven million units per year in 2008 to about five and one-half million in 2010. Thereafter, however, consumer confidence began to return and sales in 2015 may well be above seven million units again.

An important reason that new car sales increased by 36 percent between 2010 and the present is that more people had jobs and thus were more confident. The headline unemployment rate dropped from a peak of nearly ten percent in 2010 to only about five percent today. Incidentally, the seemingly perverse parallel movement of car sales and interest rates is really just the flip side of what happened during Japan’s Lost Decades. Japanese consumers were pessimistic about the future and so did not spend despite very low interest rates.

A MAC’s Impact on Interest Rates

Now we must close the circle and come back to the MAC with two questions. First, what impact will the MAC have on interest rates? Second, what impact will the MAC have on growth and jobs?

Forecasting interest rates in today’s world is a chump’s game. Not even the Fed seems to be able to decide what to do about them. And when the Fed finally decides, which it must, no one knows how markets will react because the world has not seen an increase in the Fed’s policy rate for a decade, and a lot has changed in the meantime. This said, we can make an educated guess about the impact that the MAC itself would have on interest rates. A 2006 study of the impact of foreign capital inflows on interest rates is relevant. It concluded that,
"In the hypothetical case of zero foreign accumulation of U.S. government bonds over the course of an entire year, long rates would be almost 100 basis points higher. [But] it is possible that our results overstate the effects of foreign flows.” [6]
In other words, if all capital inflows generated by the sale of U.S. government bonds were to cease, the maximum estimated impact on interest rates would be one percentage point or less. Given that net foreign holdings of U.S. government bonds account for the majority of the net increase in foreign capital in the United States on a “permanent” basis, and that net annual foreign capital inflows are the most important factor regarding increased foreign demand for dollars and dollar-based assets, this estimate of one percent should be fairly close to the maximum impact on domestic interest rates that a MAC would have if the MAC were to stop all such inflows.

However, shutting off foreign capital inflows is definitely not the MAC’s objective. The MAC only seeks to moderate net inflows by enough to restore a balance between the supply of and the demand for dollars and dollar-based assets in U.S. financial markets, a balance that would allow the dollar to return to an equilibrium rate that would balance imports and exports. The impact on domestic interest rates might well be considerably lower than one percent.

A MAC’s Impact on Jobs

The easiest way to estimate the impact of a MAC on jobs is to use the following assumptions:

  • The trade deficit is reduced from, say, 3 percent to 1 percent of GDP, a decrease of 2 percentage points or about $350 billion.
  • The trade deficit (also called negative net exports) is reduced to 1 percent of GDP by producing domestically additional import alternatives and exports with the same total value – $350 billion.[7]
  • Every billion dollars of additional net exports (which is additional GDP) will support about 6,000 jobs.[8]
  • Since the MAC could easily help decrease the trade deficit from around three percent to one percent of GDP, it would also help generate an additional two million jobs.
  • Jobs in export-oriented industries on average pay about $43,000 per year per USITA data. 
The MAC would thus help turn currently idle workers into active consumers who could afford to buy cars.


Implementing a MAC would have a minimal impact on interest rates, one that would not hurt the demand for cars. Quite to the contrary, a MAC would stimulate economic growth, employment, and the ability of workers to purchase cars and other consumer durables. The net impact of a MAC on U.S. manufacturing as a whole would be strongly positive. Consider the following:

  1. First, although a MAC might indeed increase interest rates on loans for cars and other consumer durables by about one percent because it would moderate foreign capital inflows, data going back nearly 50 years indicate that, when interest rates are increasing, auto sales normally increase as well. This reflects the fact that people buy more cars when the economy is booming, and when it is, interest rates are commonly higher than when the economy is stagnant.
  2. Second, because a MAC would make the U.S. dollar and thus U.S. manufacturing more competitive both at home and abroad, the MAC would accelerate economic growth, putting millions of workers back on the job at good wages – a process that could well lead to improved wages for existing workers as well.

In short, with a MAC in place, more workers would be in position to buy consumer durables like cars. In the longer term, restoring industry’s profitability would encourage productivity-enhancing investments in plant, equipment, worker training, R&D, and in marketing and distribution networks, leading to fundamental structural and competitive improvements in the American economy.

Having a MAC in place would also help assure that America never returns to the external deficits of about six percent that prevailed in 2005-2006 – or to a rerun of the Crash of 2008 that followed close on the heels of those disastrous deficits.

In summary, the MAC has a very high upside potential and very low downside risks.

America Needs a Competitive Dollar – Now!


  Wall Street Journal Editorial Board, 2015.11.12, “Donald Trump Is Upset.”

2   The one percent trigger level for a non-zero MAC rate has been chosen because represents a practical compromise between requiring a perfect balance between imports and exports, a degree of precision not practical in the real world, and a three percent level such as used by the Peterson Institute in calculating target exchange rate values, a level that, if used for the MAC, would de facto say that it is fine to set a public policy that leaves several million Americans needlessly unemployed. Another reason for a one percent level is that, as long as the U.S. economy is growing by at least one percent, the burden of foreign debt will not increase relative to GDP.

3  National Automobile Dealers Association, 2008.11, Automotive Financing Facts. According to Federal Reserve data, the loan to value ratio for car purchases averages 85-95 percent.

4  American Automotive Policy Council, 2014.06, State of the U.S. Automotive Industry: Investment, Innovation, Jobs and America’s Economic Competitiveness.
5 Data from the Bureau of Economic Analysis, the Bureau of Labor Statistics, and the American Automotive Policy Council/Center for Automotive Research respectively.

6  Warnock, Francis E. and Veronica Cacdac Warnock (2006.10). International Capital Flows and U.S. Interest Rates, NBER Working Paper 12560, [ Http://Www.Nber.Org/Papers/W12560]

7  Accomplishing this would probably take 3-5 years, but this figure indicates the ultimate net annual benefit of a MAC from a macroeconomic perspective.

Data from the U.S. International Trade Administration (USITA).

No comments :

Post a Comment

What do you think? Join the Conversation