In my previous post, “A high dollar means higher wages, not lower prices,” I discussed how a Canadian dollar that is well above its Purchasing Power Parity (PPP) level has led to higher wages for Canadian workers. I briefly mentioned that the C.D. Howe Institute’s Chris Ragan was correct in stating that PPP does not represent a fundamental value for the Canadian dollar. However, as pointed out by a number of readers on Twitter, I did not explain the reasoning. As a follow-up, here is why the “true” value of the Canadian dollar is not US$0.80 and, in fact, the Canadian dollar has no value that it “should be.”
A typical argument that the Canadian dollar is overvalued based on PPP goes something like this mashup of several conversations I’ve had on Twitter:
Only a market fundamentalist would believe that market prices are always right. Sometimes the market gets it wrong; think of everything from condo prices to tulip mania. Speculators/commodity prices/overseas bond investors have driven the Canadian dollar up to a level that isn’t supported by consumer prices. A 95 cent dollar is out of line with basic economic fundamentals. Look at consumer prices, which are much higher in Canada than in the U.S., as shown by an 80 cent PPP.
It is a bizarre argument for it assumes away. If we think market prices can be wrong, then why on earth would we treat PPP as necessarily correct, since it is based on a basket of market prices? PPP is a comparison of consumer price indices. The largest component of the consumer price index in Canada is shelter, which is hardly immune from the machinations of speculators. The idea that market prices “can’t be trusted” is as a pretty solid argument against using PPP as a fundamental value for the loonie.
There are a number of other hidden assumptions. This idea of PPP as the true value of a currency is based on arbitrage opportunities and is formulated in a concept known as the law of one price. Suppose the Canadian dollar is “overvalued” relative to PPP, so the Canadian dollar price of a widget is higher than the equivalent U.S. dollar price. If widgets are cheaper in the United States than in Canada, there are profits to be made in buying widgets in the U.S. and re-selling them in Canada. The buying up of widgets in the U.S. causes their price to rise. The “price” of the U.S. dollar rises as well (meaning the Canadian dollar falls), since you need the currency to buy U.S. widgets.
The dual effects of the rise in the U.S. price level and the fall in the Canadian dollar causes PPP and the exchange rate to converge. Note this is different than saying the exchange rate falls to the PPP level; convergence could happen instead by U.S. prices rising to Canadian levels and the exchange rate staying unchanged. So we need to tack on some reason as to why the price of goods shouldn’t change through arbitrage.
Beyond the assumption that consumer prices won’t change, there are all kinds of real-world frictions assumed way in this analysis. There are no transportation costs in moving the goods. Goods do not spoil in transit. There are no transaction costs when purchasing or selling the goods. There are no tariffs and no regulatory uncertainty when importing. There is no fiscal or monetary policy, or at least there are no policies that respond to changes in the currency and/or the price level.
In other words, we have to assume away all the reasons why no one in the real world is buying Big Macs in Bangalore to sell in Bergen, despite the currency deviations in the Big Mac Index. Professional economists are often criticized (and often rightly so) for the overuse of simplifying assumptions in their models. So when your analysis has enough simplifying assumptions to make an economist go pale, it’s time to re-think your model.
If the PPP level is not the fundamental value for the Canadian dollar, then what is? The truth is, the Canadian dollar has no inherent fundamental value that it should be.
Like in so many other areas, satirical website the Onion came closest to the truth when it called money a “symbolic, mutually shared illusion.” The Canadian and U.S. dollars are both fiat currencies; the greenback, for example, is not worth 22.5 grains of gold or 270 grains of silver or some other value measured in commodities (that is, not until the U.S. reverses the Crime of 1873). To illustrate, there is a well-known Milton Friedman lecture on how a 5-dollar bill is different than a random piece of paper. A version appears in Free to Choose:
Whence the difference? The printing on the $5 bill gives no answer. It simply says, “FEDERAL RESERVE NOTE / THE UNITED STATES OF AMERICA / FIVE DOLLARS” and, in smaller print, “THIS NOTE IS LEGAL TENDER FOR ALL DEBTS, PUBLIC AND PRIVATE.” Until not very many years ago, the words “WILL PROMISE TO PAY” were included between “THE UNITED STATES OF AMERICA” and “FIVE DOLLARS.” That seemed to explain the difference between the two pieces of paper. But it meant only that if you had gone to a Federal Reserve Bank and asked a teller to redeem the promise, he would have given you five identical pieces of paper except the number 1 took the place of Abraham Lincoln’s. If you had then asked the teller to pay $1 promised by one of these pieces of paper, he would have given you coins which, if you had melted them down (despite its being illegal to do so), would have sold for less than $1 as metal.
The same logic can be applied to the Canadian dollar. If the loonie has a fundamental value, it is the 75 grams of steel contained in 20 nickels. Of course, if we believed the fundamental value of the dollar is the metal contained in coins, then the Canadian dollar was massively devalued when the penny was abolished (220 grams of steel in 100 pennies). The idea that we should measure the true value of the Canadian dollar based on the steel content of the nickel is ridiculous, but it has more basis in fact than any other measure of the loonie’s fundamental value.
None of this should imply that fiat money is extraneous or unnecessary, as it serves vital functions as a medium of exchange, store of value and unit of account. Even more importantly, as Narayana Kocherlakota pointed out with the title of one of his papers, “money is memory.” As Friedman put it in Free to Choose, “Though the value of money rests on a fiction, money serves an extraordinarily useful economic function.” Furthermore, none of this should imply that the exchange rate relative to the PPP has no economic consequences. If money truly is memory, why would we assume that Canadian memory should necessarily be worth 80% of its American counterpart?
There is little value in debating what the value of the currency should be. Rather there are two questions we should be asking:
- Is the Bank of Canada’s actions appropriate given its mandate?
- Would Canada be better off if the Bank of Canada had a different mandate?
The second question is the important one, as slight changes to the Bank’s current policy stance will not lead to significant depreciation (or appreciation) of the Canadian dollar vis-à-vis the greenback. Given the value of monetary policy continuity and the associated inflation cost that would come from a mandate that would directly (or indirectly) lower the value of the dollar, there are no obvious reforms that solve more problems than they would cause.