by Edwin M. Truman, Peterson Institute for International Economics
Paper presented at a meeting of the
Study Group on North American Monetary and Financial Integration
December 13, 2002
© Peterson Institute for International Economics
I make no claim to speak for the US government, or any of its branches or agencies, on the issue of North American monetary and financial integration; in fact, the reverse. I am no longer associated with the US government. Moreover, during the 28-plus years that I was at the Federal Reserve and the Treasury Department this topic received essentially no policy-level discussion or analysis1. Any staff-level discussion was in the context of the consideration of other economic and financial unions, for example, in Europe, or tangential economic financial issues, for example, the work of John Rogers on price convergence and economic impact of political boundaries.
In contrast with Canada and Mexico, the issue of dollarization or monetary union has not been a live policy issue in the United States2. The explanation for US non-attention to this issue is the dominance of the US economy in North America. In 2001, the United States accounted for 88.4 percent of combined North American GDP with Canada at 6.2 percent and Mexico at 5.4 percent3. This fact has two important, perhaps, contradictory consequences. First, in my view, the United States, and certainly the US government, should not be perceived to be demanding monetary and financial integration; the United States faces enough tensions with its immediate neighbors with regard to concerns about political, economic and cultural "imperialism" without opening another front. Second, to the extent that Canada or Mexico was to approach the United States to explore formalizing integration in the monetary and financial area, such integration would almost certainly occur on US terms4. In other words, the North American Monetary Union (NAMU) would be based overwhelmingly on existing US institutions, the US dollar, the Federal Reserve System, and the US approach to the supervision and regulation of the financial sector, etc. because the deeper integration would be perceived as conveying few economic, financial or political benefits for the United States. It was difficult to persuade the German people to give up their beloved Deutsche mark, which had only been in existence for 50 years, in favor of the euro; it is difficult to imagine persuading citizens of the United States to give up the Greenback, which has been in existence for approximately 150 years, for another currency, despite the Greenback's checkered past as a fiat currency.
Nevertheless, the basic case for the United States to seek to formalize North American monetary and financial integration, in my view, would have to be motivated primarily by a view that in the longer term there will be and should be fewer currencies in the world and greater consistency in the structure and supervision and regulation of financial sectors around the world. In other words, a case can be made that North Americans should get out in front of these trends. However, if North Americans are to do so, the impetus, in my view, will have to come, first, from the private sector, and, second, from north and south of US borders.
The balance of this note consists of a brief review of the benefits and costs to the United States of North American monetary and financial integration, drawing upon the existing literature on this subject, plus some concluding comments.
Benefits and Costs
In what follows, I first review the benefits to the United States of North American monetary and financial integration (NAMU for short) under three headings: microeconomic, macroeconomic, and "power and prestige." I next review the costs to the United States under the first two headings.
1. Efficiency: The benefits to the United States from increases in economic efficiency as the result of NAMU are likely to be limited. If they were half a percent of GDP for the European Union, the larger relative economic size of the United States, at more than four times the largest member of the EU (Germany), suggests that the efficiency benefits of NAMU are likely to be trivial, at most a tenth of one percent of US GDP.
2. Price Transparency: These benefits, which are difficult to separate from economic efficiency benefits, are also likely to be small except for the fact that the United States has long borders with both the Canada and Mexico with a non-trivial portion of the US population living not far from these borders, which suggests that some of these border-related effects may be more than trivial.
3. Trade Expansion: NAMU no doubt would lead to an increase or a faster increase in trade than existing NAFTA arrangements because of the elimination of the inhibiting effects of exchange rate fluctuations. However, the economies of scale and scope offered to US producers and consumers are not likely to be large.
4. Financial Expansion: US financial institutions now play major roles in international finance. They also play a major role in Mexico, and there are few restrictions on their further expansion. The role of US financial in Canada is significantly smaller largely because of Canadian restrictions. However, even if these latter restrictions were removed as a result of NAMU, which I assume would be the case, the benefits to the United States and US financial institutions would be marginal at most. Perhaps, population is the best metric to use to scale the potential size of these benefits to US financial institutions. On this basis, Canada's population of about 30 million is somewhat smaller than California at 33 million5. At 97 million, Mexico's population is roughly equivalent to the US South. Thus, the benefits to US financial institutions to unfettered expansion into Canada and Mexico would be roughly equivalent to the potential benefits that have been available in recent years of expansion into California and the South to institutions from outside that state and region. The fact that one has not read a lot about such benefits suggests that the benefits of further US financial expansion into Canada and Mexico are likely to be limited.
1. Stability: It might be expected that NAMU would bring with it greater stability to the Mexican and Canadian economies as seen from the perspective of the United States, i.e., fewer economic and financial crises and, by definition, less disruptive movements in bilateral exchange rates. It is important to understand what such increased stability means from the standpoint of the United States. It could mean less fiscal burden in terms of assistance, but the direct burdens have been minimal in recent years for the United States, even in the case of Mexican crises. It could mean less macroeconomic volatility in the sense of a reduced impact of economic fluctuations in Canada and Mexico on the rest of NAMU. However, this is likely to be a small effect. Canada is about the economic size of Texas, based on real gross state product (GSP) in 2000 adjusted to a current dollar basis; Mexico is roughly the economic size of Florida and Alabama. Economic fluctuations in these states do not have large impacts on the rest of the United States6. Moreover, the issue with respect to Canada and Mexico, from a US perspective, would be the implications of confining more of the burden of adjustment from disturbances in those countries to those countries rather than exporting it to the United States, inter alia, via exchange rate fluctuations. US goods exports to Canada and Mexico are 1-1/2 and 1 percent of US GDP, respectively. If each of those economies were to decline by 5 percentage points of GDP relative to their trend, the impact on the US economy would be between 0.075 and 0.05 of a percentage point of US GDP, assuming an income elasticity of demand for imports of one. Even if those figures were doubled due to exchange rate effects, the impact would be tiny. Principally the exchange rate effects would be bottled up in Canada and Mexico as a result of NAMU, which would be a non-trivial amount, but not large. To be sure, the consequences to Canada and Mexico of the bottling up of their disturbances might be deeper or more protracted economic declines, which would increase the negative direct income effects of declines in Canada or Mexico on the United States.
2. Growth: The potential positive growth effects on the United States of NAMU derive from increased trade and increased economic efficiency. No doubt they would be present, but as noted above under microeconomic benefits, the first-round effects are likely to be quite small, for the United States, and the second-round effects even smaller.
3. Monetary Framework: Both the Bank of Mexico and the Bank of Canada now employ inflation targeting as their framework for the implementation of monetary policy. Critics in Canada, in particular, have pointed to the Bank of Canada's success over the past decade in reducing Canada's inflation rate below that of the United States, although for most of the post-World War II period Canada's inflation rate was above that in the United States and inflation in the 1990s in all three countries has been substantially lower than in the 1970s and 1980s7. I personally would favor the Federal Reserve's adoption of inflation targeting as its framework for monetary policy, but any expected gains in terms of improved macroeconomic performance, compared with the reasonable assumption of an extension of the average performance of the past decade, are likely to be marginal.
Power and Prestige
1. Some advocates of NAMU have argued that from the standpoint of the United States a broader US dollar area would enhance the role of the dollar in the global financial system and protect it from inroads from the euro, for example.
2. For my money, double meaning intended, the net benefits to the United States of the dollar's large role in the international financial system are between small and negligible. Any benefits accrue to US financial institutions because they are US dollar-based and world commerce operates largely in US dollars. Offsetting these benefits are costs in terms of constraints on US monetary policy, largely as the result of feedback effects on the US economy of US monetary policy actions and the tendency of the exchange-rate channel for monetary policy to be blunted. Moreover, I see no threat to the dollar's role from the euro or any other currency in the near future.
3. Thus, any benefits from an enhanced role for the dollar in global finance as a result of NAMU are more likely to accrue to the international financial system as a whole in line with my comment above about the benefits from a generalized move toward fewer currencies and a more homogeneous international financial system.
1. Switching: Monetary integration scorekeepers often point to "switching costs" as a fixed cost offset to the dynamic microeconomic benefits of monetary union. From the standpoint of the United States, and under the assumption that it would be Canada and Mexico that would be switching to the use of the US dollar rather than the three countries switching to the use of a new fourth currency, there would be little or no switching costs for the United States from NAMU, aside from the costs borne by US firms and financial institutions operating in Canada and Mexico. Indeed, the avoidance of such costs would be an advantage to the three participants from using the currency of the country that today accounts for almost 90 percent of all transactions in the region.
2. Supervision and Regulation: A minimal necessity of NAMU, in my view, would be the alignment of supervision and regulation of financial systems in the three countries, at least with respect to lender-of-last-resort policies and procedures. As noted earlier, I would expect that this alignment would be toward if not in full congruence with the US model. In the process, there may be transitional costs imposed upon the United States and US financial institutions. It is also possible that the resulting system could be an improvement over the current regimes in the United States, as well as those in Mexico and Canada, converting a net cost into a net benefit.
1. Monetary Policy: I would expect that the addition of Canada and Mexico to the United States as a monetary area operating with one monetary policy would produce essentially no change in US monetary policy, and therefore have little impact on the United States economy compared with the absence of NAMU. As noted above, in economic terms adding Canada to the US economy would be like adding Texas again to the US economy, and adding Mexico would be like adding Florida and Alabama to the US economy. The resulting monetary policy is not likely to differ significantly. To appreciate the essential truth of this comparison, imagine subtracting those three states from the 50 states that now make up the US monetary union. It is difficult to believe that there would be any substantial difference in US monetary policy going forward8.
2. Fiscal Policy: It would be for consideration by the participants in NAMU whether there would be parallel fiscal arrangements. One could imagine that there might be some, in particular in support of Mexico. These might be designed to accelerate the growth of income per capita in Mexico to reach the US level; there also might be a greater emphasis on the development of cross-border infrastructure. Whether such arrangements would extend to counter-cyclical fiscal policies and what the costs might be to the United States is a matter for speculation on which even I in this highly speculative note am not willing to opine.
3. Exchange Rate Inflexibility: Earlier I listed as a potential benefit of NAMU for the United States the reduced scope for the Mexican and Canadian economies to export a portion of their economic fluctuations to the United States, inter alia, via the exchange rate channel. However, on the cost side one could imagine that the reduced exchange rate flexibility for the US dollar as a consequence of NAMU could impose added adjustment costs on the United States. Although these costs are likely to be small in the current context, if the Mexican economy were to expand to a size commensurate with its relative population today, roughly a quarter of the North American total, one could imagine that the United States might wish it could devalue the US "dollar" relative to the Mexican "dollar," as was the case with respect to Japan and Europe in the early 1970s when the United States was trying to extricate itself from the constraints of the Bretton Woods exchange-rate system. In the limit, one might imagine that NAMU might be reversed, and that would most likely impose substantial economic and financial costs on the United States.
4. Seigniorage: Some arrangement would have to be made for the sharing of seigniorage in the context of NAMU9. This is probably one area where there might be a small net cost to the United States if for no other reason than the formula would likely be based on a fixed proportion of US dollars in circulation at the time of union. If that amount continued to grow at a faster rate than the growth rate for the actual use of currency within NAMU because of rising demand from outside the union, Canada and Mexico would benefit at the expense of the United States. In addition, if the initial formula were based on GDP shares or tried to take account of the fact that the US dollar circulates today in Canada and Mexico to a limited extent, then the formula would also advantage these countries and disadvantage the United States compared with the status quo.
1. The above review of the costs and benefits to the United States of North American monetary and financial integration suggests that neither costs nor benefits are likely to be substantial, and the net is likely to be trivial as seen from narrow perspective of US short-term economic interests.
2. As a consequence the case for or against such integration is likely to turn on longer-term economic and financial considerations, for example, whether the global financial system is likely to function better with fewer currencies, as well as on political considerations.
3. Explicit in what I have said is the view that North American monetary and financial integration implies substantial convergence in the structure of financial systems and in supervisory and regulatory regimes, preferably more convergence than has been the case to date in the European Union. Although the US financial system and regulatory and supervisory regime is far from ideal, my operating assumption is that the North American financial system of the future would look much like the US financial system of today, and likewise for monetary institutions.
4. Implicit in what I have said, and certainly what I believe, is that for North American monetary and financial integration to be successful, it would have to involve an elimination of all barriers to labor migration. This is likely to be seen as an economic necessity and a political difficulty.
1. I did testify on February 8, 2000, before the Economic Policy Subcommittee of the US Senate Banking Committee on the issue of dollarization and Senator Mack's proposed legislation to establish a framework for potentially sharing seigniorage with countries that decide to dollarize, but the focus of that legislation was not on North America.
2. The only exception to this generalization of which I am aware is the occasional reference to the need for a macroeconomic or monetary chapter as a part of or an enhancement to NAFTA; see, for example, Robert A. Pastor's Toward a North American Community. Most of that commentary has focused on the adverse effects on the US economy of depreciation or devaluation of the Canadian dollar or Mexican peso, for example, in the context of the Mexican financial crisis in 1994-95.
3. On a purchasing-power-parity basis, with the United States as the base, the US share drops to 84.9 percent, while Canada was at 7.4 percent and Mexico at 7.7 percent. In terms of population, a measure of potential relative size over the very long term, the US share drops further to 67.5 percent, with Canada at 7.7 percent and Mexico at 24.9 percent.
4. I think "formalizing" is the correct term because I expect there to be enhancements of integration, at least in the financial area, without entering into further formal arrangements involving the three countries.
7. Over the past 10 years the annual increase in the Canadian CPI has be 1.5 percent, 8 tenths below the US average for a cumulative difference in levels of 4 percent. However, over the past 30 years the cumulative difference in levels "in favor" of the United States has been 6 percent.
8. Note that the economic size of the three states is essentially the same as the economic size of California. California's economic downturn in the early 1990s, which triggered substantial net out migration from California, had no discernible impact on Federal Reserve policy at the time.