John Quiggin had a good piece in the NYT, pointing out how the sky-high valuations of Bitcoin undermine the efficient market hypothesis that plays a central role in much economic theory. In the strong form, we can count on markets to direct capital to its best possible uses. This means that government interventions of various types will lead to a less efficient allocation of capital and therefore slower economic growth.

Quiggin points out that this view is hard to reconcile with the dot-com bubble of the late 1990s and the housing bubble of the last decade. Massive amounts of capital were clearly directed towards poor uses in the form of companies that would never make a profit in the 1990s and houses that never should have been built in the last decade.

But Bitcoin takes this a step further. Bitcoin has no use. It makes no sense as currency and it is almost impossible to envision a scenario in which it would in the future. It has no aesthetic value, like a great painting or even a colorful stock certificate. It is literally nothing and worth nothing. Nonetheless, at its peak, the capitalization of Bitcoin was more than $300 billion. This suggests some heavy-duty inefficiency in the market.

Quiggin is on the money in his analysis of Bitcoin and its meaning for the efficient market hypothesis, but it is worth taking this line of thinking in a slightly different direction. The purpose of the financial sector is to allocate capital. In principle, we would want as small a financial sector as possible, just like we would want a small trucking sector.

Both sectors are providing intermediate services. While they are both essential for the operation of the economy, they do not directly provide benefits to people, like health care, education, and housing. In general, we think more of these and other final goods and services are better, but we want to have as few resources (labor and capital) tied up in finance and trucking as possible.

We have seen the opposite story with the financial industry over the last four decades. If we go back to the mid-1970s, the narrow financial industry (securities and commodities trading and investment banking) accounted for a bit more than 0.5 percent of the economy. It has nearly quintupled relative to the size of the economy, as it now accounts for more than 2.3 percent of GDP. The difference of 1.8 percentage points of GDP is almost $360 billion annually in today's economy. This should be a cause for serious concern.

We currently have a bit less than 1.5 million workers employed in the trucking industry. Suppose that the industry were more than four times as large and instead employed 6 million workers. This would be a huge drain on the rest of the economy since we would have to pay for the salaries, trucks, and fuel for four times as many workers.

If we had something to show for these additional trucks and drivers then we might decide the additional cost was worth it. If it meant, for example, that we had less food spoil in transit or that we were all getting the goodies we ordered online within minutes after we clicked the purchase button, then perhaps the additional expense from this much larger trucking industry would be reasonable. But suppose we had four times as many trucks and truckers and our service was no better than it had been before.

Arguably, this is the story of the financial industry. Is there any reason to believe that it has done a better job of allocating capital to its best uses in the last two decades than it did back more than forty years ago? Sure, some innovative companies have gotten startup capital and changed the world, but that was true fifty years ago as well. Just at the most basic level, productivity growth was much more rapid in the 1950s and 1960s, when we were devoting a much smaller share of our resources to the financial sector than is the case today.

This is why I am such a big fan of a financial transactions tax (discussed in chapter 4 of Rigged: How Globlalization and the Rules of the Modern Economy Were Structured to Make the Rich Richer [it's free]). Even a modest financial transactions tax (FTT) would effectively take a sledgehammer to the financial industry. A tax of 0.2 percent (20 cents on one hundred dollars) on stock trades, and scaled for other financial instruments, could plausibly cut the size of the narrow financial sector in half, freeing close to $200 billion a year for productive purposes. (That's more than $600 per year for every person in the country.) In addition, this would be a huge blow against inequality since many of the richest people in the country get their income from the financial industry.

Anyhow, that's the economics of an FTT and it follows pretty directly from the demolition of the efficient market hypothesis. If we can't count on a larger market and more transactions to move us to a better allocation of capital, then let's look to make the sector smaller and stop wasting resources that accomplish nothing. This would essentially mean fewer stock trades and fewer complex financial instruments. That would be a better world.