Ignoring transaction costs would be perilous

At the end of the first year of graduate school in economics, they make you take long tests on micro- and macroeconomics. My micro test lasted five hours, and I still remember a question I missed.

It was about transaction costs: it asked what happened if buyers and sellers had to row from island to island in order to exchange their goods. It turned out that this so-called transaction cost, even though it was tiny, made the market break down — no one bought or sold anything.

As I learnt that day, transaction costs are very important — so important that they change our basic understanding of economics. But many economists ignore them when creating their models. And the basic mental frameworks that most of us use to understand the economy — for instance, good old supply and demand — leave out transaction costs.

Financial markets are a good example. As any successful trader knows, lots of trading strategies that look successful at first glance end up losing money once transaction costs are taken into account. Many a first-time trader has been seduced by the promise of a sure moneymaker, only to find that costs like price impact and slippage (which are ways the market moves against traders) can turn a winning trade into a loser.

Financial economists, for their part, have finally started to incorporate trading costs into their models, with interesting results.

Transaction costs are also fundamental to the reason why companies exist in the first place. It’s fun to imagine a world of backyard capitalism, where everything is outsourced, everyone is a contractor and people simply come together spontaneously to produce any good or service needed.

Many economic models work like this. But it’s obviously a fantasy.

In the 1930s, economist Ronald Coase theorised that transaction costs are the reason we have companies. The theory is simple and intuitive. Suppose you’re a manager who wants to create a new feature for your app. Instead of having to hunt down a team of programmers — which would cost time and money — you use the ones you already have.

Obviously, the internet has changed this game. It’s easier to find and communicate with people outside one’s own company now, which is a large part of the reason we’ve seen a boom in supply chaining, offshoring, consulting and outsourcing of all kinds.

But even in the online age, transaction costs aren’t zero; it still takes effort, time and money to verify the trustworthiness and quality of a supplier or business partner.

Some companies even found that they took the outsourcing boom too far, and decided to bring some operations back in-house.

Transaction costs can also be very important for economic policy. Simple theories of free trade — such as the theory of comparative advantage — assume that companies and workers can adjust quickly and easily to the industry shifts caused by a trade agreement.

But in reality, workers can find it difficult to change careers or find new jobs. It can require moving to a new town, going back to school, losing one’s work friends and many other kinds of disruption. Companies too may find it hard to adapt their business models.

This is probably the reason the surge in trade with China, after that country’s entry into the World Trade Organisation, caused such disruption in the US. Whereas earlier trade booms — with Japan, Europe, or Southeast Asia — were limited enough in size and scope to allow laid-off workers to find new jobs similar to their old ones, the China shock was so big and comprehensive that many workers never recovered. The transaction costs were just too high.

Transaction costs can also be important in housing markets. When a landlord raises rent or evicts a tenant, the landlord pays the cost of finding a new occupant.

But the tenant pays a much steeper cost — he or she must now search for a new place to live, a new route to work and new places to shop, in addition to paying the actual cost of moving.

This asymmetry in transaction costs is one reason people feel landlords have unfair power over tenants, and support policies like rent control.

So, a world with transaction costs can look much different than one where markets adjust quickly and frictionlessly. Just like on my graduate school exam, even a little bit of sand in the gears of free exchange can make markets malfunction.

Economists should probably incorporate transaction costs into more of their models, but even more importantly, we economic writers and pundits shouldn’t forget them when we talk about policies.

Leave a Reply