The most fundamental aspects of the Canadian and U.S. economies have not diverged in the past half-century. So why has our dollar been falling since 1950? And, more to the point, how do we stop it?

Remember that consensus forecast for a strengthening of the Canadian dollar last year? Indeed, remember the consensus forecast for a strengthening of the Canadian dollar in ... you pick the year? The dollar has consistently defied all official predictions about its value from all the usual technical sources. Why has all this optimism been consistently dashed? What is really going on?First, let''s look at what has happened to the dollar since 1950.

As Graph I shows, while there have been fluctuations -- most associated with temporary circumstances -- the predominant feature of the Canadian dollar exchange rate against the U.S. dollar is that it has declined. Why?Forecasters have used several factors to explain the decline and predict the future of the dollar: commodity prices, interest rates, government borrowing and, perennially, purchasing power parity. But do these standard theories really explain the 50-year performance of the dollar?

Let''s begin with commodities. The leading proponent of the commodity argument is the Bank of Canada. Graph II shows the Commodity Research Board''s commodity price index (adjusted for inflation) since 1950. The trend is certainly in the right direction. If we are getting less for our commodities and paying more for the things we import -- if the so-called terms of trade are against us -- that would explain the downward drift in the dollar. But the fact is that the importance of the terms of trade has been shrinking throughout the period as Canada has increasingly become an importer as well as exporter of commodities.

As recently as 1976, imports of $6.9- billion were equal to 56% of exports valued at $12.3-billion. In 1999, commodity imports had risen to $82.5-billion and represented 75% of our commodity exports. In other words, it has become increasingly irrelevant that we get fewer manufactured goods for our commodities because we also have to pay less for the ones we buy. If our commodity trade were balanced, any change in their prices would be a wash and therefore unable to explain any changes in the exchange rate.

Even more damaging for the commodity explanation is the fact that the Bank of Canada''s calculation of the overall terms of trade -- the price of everything we export compared with everything we import -- shows that there has been an improvement of either 1.6% or 3.5% (depending on the form of index) since 1992, when the dollar had the now seemingly stratospheric value of 77¢. Why hasn''t the dollar risen instead of fallen as the overall terms of trade improved?

There is still a role left for commodity prices, however, especially in predicting short-term changes as some forecasters, such as the Bank of Canada, do quite well. That is, to the extent that currency traders still view Canada as primarily a commodity play, they might well trade the currency on changes in commodity prices. This would produce the sort of short-term correlation which the Bank''s forecasters find -- in spite of the lack of a credible structural connection between commodity prices and the trend value of the currency.

Some studies, based on purchasing power parity, have attempted to show that the drop in the value of the Canadian dollar is temporary because the development of the cost of living in the two countries favours Canada. Eventually, so the story goes, goods arbitrage may be expected to correct this mismatch and restore purchasing power parity through a higher Canadian dollar. While this reasoning may have a kernel of truth, purchasing power parity calculations have been an even more miserable predictor of the exchange rate than the terms of trade. Moreover, inflation measured through consumer price indices in the two countries has been nearly identical during the period (Graph III). Certainly there is no trend here that would explain the currency moves.

Interest rate differentials with the United States are often used to explain movements in the exchange rate on the grounds that they drive the behavior of portfolio managers concerned with short-term profits. As Graph IV shows, while interest differentials have fluctuated, there is no secular trend in them that would explain the long-run decline of the dollar.

So what has determined the dollar''s decline? What are the underlying economic conditions in Canada and the United States which might explain the ongoing, persistent underperformance of the dollar? To solve the puzzle we have to look for something that has evolved differently in the two economies since the 1950s when the currencies traded about one for one.

It is clear that the most fundamental characteristics of the economies in the two countries have not diverged. In fact, they have become more similar. Our autos are now the same and we produce a lot of them for both countries. Our service sector is a mirror image of the United States. Our fast-food restaurants, auto service and general retailers are the same. Most of the huge flow of trade between the countries is intra-firm trade. As noted, even our commodity trade is headed toward balance.

The main way in which the economies of the two countries have diverged is in the size of the public sector and the average tax rates as reflected in total government revenue as a percentage of national income. As Graph V shows, in 1950 the two countries had proportionally identical public sectors and roughly the same total tax burdens. During the late 1990s, tax burdens in Canada were 50% greater than in the United States.

I would like to propose that it is this simple fact -- the divergence in the size of the tax wedge in the two countries -- which explains the decline of the Canadian dollar by roughly the same percentage during the last 50 years.

How does an increase in the general tax rate affect the exchange rate?Harry Johnson and Mel Kraus, in a 1970 article in the Canadian Journal of Economics, Border Taxes, Border Tax Adjustments, Comparative Advantage, and the Balance of Payments, had an explanation. It would come from the reshuffling of wages, prices and rates of return as people try to avoid, pass on or simply adjust to the higher taxes. In the end, some combination of reduction in wages, rates of return or the exchange rate is required to account for the burden of the tax since for all participants it cannot be avoided.

What does that bafflegab really mean?Take, for example, an additional tax imposed on wages. The intent of the government is to divert to itself some of the purchasing power which has been earned by workers. The employer must remit an increased portion of the workers'' pay to the government. Workers attempt to compensate by demanding higher wages, employers try to pass it on in higher prices.

As we have seen in the comparative inflation data, monetary policy has prevented employers from doing this on domestically targeted production. Canadian inflation has not been very different from U.S. inflation. Since not all employers can pass on the wage demands in higher prices, they resist the demands. In the end, most of the burden is borne by workers in the form of wages which are lower than they would otherwise have been. The decay in purchasing power of workers, particularly in the last decade, has been widely documented.

However, domestic monetary policy does not affect the prices which can be charged on exports. What does affect them is the state of U.S. demand for the products involved. To the extent that exporters try to raise their prices against the prices set in the U.S. market, the demand for Canadian exports falls. Given no change in the level of imports, this will put pressure on the value of the currency until it has fallen to the point where the attempted passing-on of the tax increase is just offset by the reduction in the exchange rate.

The foregoing tendencies are amplified by the fact that any rise in Canadian prices is an opportunity for imports to take a larger share of our market. Relatively more imports mean relatively more downward pressure on the dollar. During the 50 years we are considering, the progressive liberalization of trade policy has opened more and more of the Canadian market to this effect. Accordingly, we have become more sensitive to the exchange rate impact of tax increases.

Of course, this process is masked by the fact that wages and prices are constantly changing. Moreover, it is important to remember that we are talking about the comparative and differential impact of taxation in two economies. If Canadian and U.S. governments raised identical taxes, there would have been no loss in competitiveness.

As for the impact of tax increases on capital, perhaps no special explanation is necessary. Capital which is free to move shifts at the margin to minimize tax incidence. As it moves out, it puts downward pressure on the value of the dollar.

While the Department of Finance and the Bank of Canada might cringe to be associated with the theory of exchange rate movement projected here, the Department of Finance used essentially the same line of argument when it was selling the GST to Canadians. One of the advantages, they said, was that, as a value-added tax, it could be removed from exports and hence boost the level of Canadian exports. So, from any given starting point, removing tax boosts exports and the value of the currency.

The Department of Finance did not go on to say that the reverse is also true, but we have seen that it is.The future development of the exchange rate trend depends on whether the tax gap between Canada and the United States will be closed. While the government''s plans for tax cuts in Canada promise that it will be narrowed, these cuts may well be offset by possibly even larger cuts by the new Bush administration.

I predict that during the next few years the Canadian dollar will have its usual short-term fluctuations determined by interest rate differentials and developments in the commodity markets, but that its trend will be determined by the relative tax burdens in the two countries. If my analysis is correct, the trend in the value of the loonie will recover only if we cut taxes more than the Americans.