Why Institutions Matter

Serious economists talk a lot about institutions, and how they matter to well-being.

Not-so-serious economists — and in this group I’ll include politicians, bureaucrats, and most of the media and pundits — don’t talk about institutions much at all.

Why is that? What the heck are institutions? How could institutions make a difference to well-being? And how could some people “get it” while others do not?

Don Boudreaux has an absolutely stunning post at Café Hayek that may fill in some of the gaps in your thinking. I’m going to repost the entire thing here (sorry Don!).

Don’s post is a two-parter. You might find the first part a little boring and long-winded. Try not to. And I recommend re-reading it after you’ve read the second part, because it might make more sense then.

What Don is saying in the first part is that too many people, especially in positions of power, view well-being derived from economic outcomes as the result of solving some sort of engineering problem.

What the heck does he mean by that?

I’ll give you an example. Think of your smartphone, or the computer you use the most, or your home theatre system, or your car, or your filing or organizational system, or maybe your household’s financial situation and plan for the future. Got your example in mind? OK, then think of someone else managing their version of the same thing: their smartphone, their computer, their home theatre, car, their filing or organizational system, or their financial situation. And let’s think about those people: do you have friends that think if they just make one more tweak to that thing that their world will be all better?

My guess is that some of you are laughing to yourselves a little. Because we all know people like that. And I added the emphasis to “all” because I wanted you to think of that in the way that a parent might say it to a preschooler: “I’ll fix your boo-boo and make it all better”.

People who think this way (and we all do, just some to a greater extant than others)* are looking for an engineering solution to the problem: if we just keep tweaking it, eventually we’ll get it right.

And, unfortunately, many people double-down on that bit of bizarre human behavior by adding in some magical thinking. Something like, if I can just get my smartphone, or computer, or home theatre, or car, or filing system, or financial situation right, then everything else will fall into place. You know … people I find attractive will start finding me attractive first, and my dog will start listening to my commands, and their will always be a good parking spot when needed.

I’m not saying this is bad (and neither is Boudreaux). But we are saying that this is 1) very human, and 2) not always appropriate.

For example, football is the ultimate game for the engineering approach: endless drills on fixed patterns of play so that it will all work if everyone does their job at the same time. Except football teams can’t do that very successfully because there are always intangibles like team chemistry and home field advantage getting in the way. (WTF is home field advantage anyway? We all know it’s there, but WTF is it? Something in the water? No one knows … and that’s my point). So for a football coach, it’s very human to think that they can plan everything out, but not always appropriate to think that planning can produce the winning team they desire.

So, the first part of Boudreaux’s discussion is about what happens when people view the economy as an engineering problem:

Joan McDonald, a home-schooling mom, writes to ask if I can explain what we economists mean when we say that “institutions matter.”  I’ll give it a shot.

First of all, not all economists appreciate the role of institutions.  Economists who most ardently insist that institutions matter typical do so in order to counter the impression left by other economists – as well as by the many other people who are keen on guiding the economy less through market forces and more with central direction – that the economic problem is merely, or mainly, an engineering one.  Given all people’s tastes and preferences – given some known existing stock of capital, resources, and inventories – and given the current state of technology, the manner in which capital, resources (including labor), and inventories can be allocated in order to satisfy as many consumer desires as possible can be described mathematically.

Of course, the inconceivably immense number of such variables in reality means that no actual mathematical description – one with real quantities throughout – can in practice be written down.  What can be written down, though, is a description of the logic by which all the actual and given resources, inventories, and bits of capital must be allocated in order that the result is the satisfaction of as many as possible of the existing and given consumer preferences.

Such mathematical (or geographical, or even plain-word) depictions of the logic of ‘the economic problem’ makes this problem appear to be an engineering one.  Such depictions make central direction of the economy appear to be not only a plausible, but a preferred, method of ‘solving’ the economic problem.

Although there’s nothing inevitable in such engineering depictions of the economic problem that turns analysts’ attention from the role and reality of change, uncertainty, discovery, dispersed knowledge, and – most importantly – incentives, there seems to be something psychological in such depictions that has this attention-diverting effect on those who obsess over getting correct the formal mathematical description of the economic problem.

Those who ‘see’ the economic problem chiefly as an engineering challenge typically don’t think deeply enough about the available real-world conditions that can with the greatest likelihood of consistent success not only (1) approximate the best solution that can in principle be described mathematically with a set of given variables, but also (2) encourage and accommodate real and beneficial changes in the variables – changes such as occur, in the real world, in consumers’ preferences and in production techniques.  “Institutions” is the summary term for the set of real-world conditions that in fact govern how existing resources are allocated and the extent to which change is encouraged and accommodated.

Institutions can be “formal” – such as a particular government’s tax policy and occupational-licensing dictates.  Most institutions, however, are informal.  These are the rules of behavior that people generally obey even though these rules are not written down in the sovereign’s rule book and imposed formally.  They include the vast and intricate ‘constraints’ and expectations that are the evolved law and social customs and norms.  Some of these informal institutions reflect purely cultural happenstance (say, the fact that people today typically greet each other by shaking hands), while others reflect hard-wired human nature (such as the fact that each of us cares more about ourselves, our loved ones, and our friends than we care about strangers).

These formal and (mostly) informal ‘rules’ governing behavior combine with each other to influence the ways that each person in the multitudes of people in society act.  Some patterns of actions will produce ‘good’ overall economic outcomes; others will produce ‘bad’ overall economic outcomes – regardless of the formal details of the mathematical description of how resources can be ‘best’ allocated.  In other words, what actually happens in an economy is the consequence of institutions and not of the engineering problem that the economy faces.  The challenge – and it is a monumental one – is to get the institutions right; the challenge is not to describe in principle how best to allocate resources.  The latter task is child’s play in comparison to the former.

A good contemporary example of all this is Obamacare. A huge healthcare law about which the leader of the Democrats trying to pass it through Congress remarked “We have to pass the bill so you can find out what is in it.” That’s the engineering part. Now add a dash of magical thinking: we never passed this before because Republicans (who probably need to be wearing black hats in case people don’t get the message) wouldn’t let it pass, and Republicans continue to be the biggest threat to its success.

Of course, if you don’t like that example, we could use the 2003 invasion of Iraq and the plan for “nation building” put in place by the second Bush administration as the engineering solution. Then the magical thinking part is that the only way it could succeed was to keep the Democrats from getting elected and goofing it up.

But Boudreaux’s point is that this sort of thinking usually doesn’t work out well because institutions are more important to economics outcomes. Here’s the second part:

Pardon me if the above is unclear or too airy-fairy.  Let me end with a simple thought experiment.

Suppose that the rules of the economic ‘game’ are changed so that each household in America gets to spend, not the income that it earns, but only the income earned for it by some other household.  Each household is paired at random with another; no paired households know each other; they remain strangers to each other.  Let this be the only change in America’s institutions and economic landscape.

What do you predict will happen?  I predict that productive activity outside of each household will fall dramatically.  People will work fewer hours earning incomes at their jobs; they’ll invest far less in businesses and in financial markets; and they will display far lower levels of entrepreneurship in the market.  In turn, people over time will become much poorer materially.  People also will likely become less cosmopolitan as their engagement in commercial markets declines.  I predict also that there will be lower levels of formal schooling.

Note that in this hypothetical example the formal description of the ‘best’ way to ‘solve’ the economic problem isn’t changed.  People’s tastes and preferences are unchanged, and at the start of this social experiment America’s stock of resources, capital, and inventories is the same as it would have been had this social experiment not been launched.  Yet make this one change – which is an institutional and not an engineering one – and the economy’s performance changes greatly.

Of course, most institutional changes aren’t as dramatic as the one hypothesized here.  Therefore, the results of changes in institutions – good and bad – are more difficult to detect and to distinguish from other changes.  But even small changes institutions, we can be sure, affect the way the economy performs.

I think that example is almost perfect. What I would add is that the example needs to point out that while each family gets someone else’s production (because it’s transferable) they get to keep their own leisure (which isn’t). That’s really the key.

When we look around the world and try to explain why some regions are rich and others are not, capital accumulation and resources only go so far: the scale of differences in capital and resources needed to explain the differences we see in well-being is implausibly large, and growth models suggest those differences should disappear over the course of decades rather than centuries.

So that leaves us with technology, and to be sure there are large differences in technology between the rich and the poor. The thing is, there’s little that impedes the free flow of the “high tech” that we normally associate with technological advancement: your smartphone doesn’t cease to work when you cross a border, and so its value doesn’t change much either. If there aren’t actually large differences in technology between rich and poor, then those differences will hardly make a decent explanation for why we end up with differences in well-being.

Instead, we need to envision “low tech”. This is the sort of stuff that is a technology, in the sense that it doesn’t depreciate and it’s non-rivalrous. But, it stops or slows down when it needs to cross borders. The name we give to this is institutions: the ways of doing things that influence the efficient use of more capital by our labor to be more productive. That’s the thing that’s left to explain the big differences in well-being that we observe.

But people don’t like to talk about low tech institutions too much. They only change incrementally and slowly. Which means that a lot of times we’re stuck with them for the long haul. And because we’re stuck with them, we get attached to them as well. Think about how many people in human history have tried desperately to hold on to the institution of slavery, and yet it clearly doesn’t work as well as other economic systems.

But, because what I referred to at the top of the page as “non-serious economists” don’t talk about institutions enough, many people are flying blind when they try to figure out why the world’s economies are the way that they are.

* Just a note here about what college is supposed to do for you. I wasn’t sure if I should use “extent” or “extant” in that sentence. So I have this funny thing I do: I looked it up on The Google. You should get in the habit of doing this too; if you learn one thing in college that helps you in the real world, the willingness to ask about something you’re not sure of so that you do it right once instead of doing it twice might just be it. And if you’re curious, I was leaning 60/40 towards “extent”, and I was wrong. Good thing I checked.