Constrained Policy Recommendations

The role of economics in policy discussions

You are reading Economic Forces, a free weekly newsletter on economics, especially price theory, without the politics. Economic Forces arrives weekly in the inboxes of over 8,000 subscribers. You can support our newsletter by sharing this free post or becoming a paid subscriber:

---

Economists are trained to make purely positive, scientific claims: what happened to GDP? what will happen if we change the tax rate? We aren’t perfect, but that is where the science shines.

At the same time, economists (myself included) often get involved in policy debates outside of positive economics. On the one hand, we have no special standing on what policy should be. On the other hand, I will argue that our scientific knowledge can help guide policy toward specific ends. Will your proposed means achieve your stated goals? Many economists lump these both under “normative economics”—in contrast to positive economics—but, as I’ve written before, they are different approaches.

What is Optimal Policy?

From one angle, we already have the best feasible policies in place. Just as consumers pick their preferred bundle when shopping and markets exhaust gains from trade, politics exhaust gains from political trades. If sugar tariffs exist because the farm lobby wants it, there must not be a cheaper way to buy off the farmers with political power. Yes, we could wish farmers were not self-interested and didn’t use the political process, but that is on par with wishing candy fell from the sky. In the real world, the farm lobby exists and subject to that, this policy is the efficient.

Is Politics Efficient? A discussion of Filip Palda's "A Better Kind of Violence"

This insight is behind the great Armen Alchian’s idea that “whatever is, is efficient.” It’s an important lesson to learn so we remain humble in our ideas.

But taking this view too seriously would mean that we give up on any hope of making improvements. We wouldn’t even act. You can’t do nothing about nothing! After all, if there were a way to make things better that was worth the cost, it would have already been done.

Recognizing Constraints

Instead of assuming that everything is perfect, we can acknowledge that there are always trade-offs to be made. This approach is encapsulated in Milton Friedman’s refrain “There ain’t no such thing as a free lunch,“ or just TANSTAAFL.

Whenever we make a policy recommendation, we implicitly or explicitly recognize that there are costs associated with it. We might want Policy X, but we know it will come at a cost of Y. All policy recommendations involve some form of constraints, and while we can debate whether Policy X is good, we should always be aware of the costs and limitations involved.

Economists are great at recognizing constraints. That’s our superpower. For instance, when someone argues that "Everyone should have free health care," economists rightly highlight that resources are limited and that providing health care to one person means taking resources from somewhere else. Health care doesn't just appear out of thin air, nor does it come from a benevolent government agent. This contrasts an unconstrained (Nirvana) policy recommendation that assumes that health care is free and ignores the existing real-world constraints.

Explicit economic models offer a unique advantage in forcing us to recognize the constraints present in the economy explicitly. They force the users to do a proper accounting. For example, we can't consume more apples than what is produced, and supply must always equal demand. Through practice with economics and economic modeling, this way of thinking becomes second nature for economists. By understanding the economic constraints, we can better appreciate the trade-offs involved in making policy decisions and avoid unrealistic assumptions that ignore the complexities of the real world.

The constraints we face in the world extend beyond just resources. Information is limited, and acquiring knowledge can be costly. Additionally, there are policy constraints to consider, such as legal, political, and procedural limitations. We can’t imagine away the corn lobby. Any policy recommendation takes an implicit stance on these constraints and what is feasible, and it’s not always clear how we should think about them. In the world we have with the politicians we have facing the constraints we have, is the recommended policy the best we can do?

The benefit of economic training in thinking through means-ends analysis is that economists are trained in applied constrained optimization problems. For all its flaws in terms of economics education, spending lots of time on constrained maximization means this is second nature to the well-trained economist. We can think through problems of the type “Given some goal and some set of constraints on what is feasible, does the proposed policy achieve the goal?” While economists are comfortable with the mathematical distinction between constrained and unconstrained choices, the general insight is beautifully expanded on in Thomas Sowell’s classic A Conflict of Visions. In Sowell’s book, he distinguishes between constrained and unconstrained visions. Someone who has a consistent view of the world as constrained will also make constrained policy recommendations, as I am arguing they should.

Room for Improvement

However, we can still recognize that there are trade-offs to be made and that our current policies may not be optimal in some other sense. Just as there are always gains from trade remaining in markets, there are gains from trade still in politics. Acknowledging this allows us to look for ways to improve our world within existing constraints.

The reason in both markets and politics is that constraints may not be binding as we imagine. I find an analogy to consultants in the private sector helpful. After college, I worked on a software implementation at a manufacturing plant. We brought in consultants, but the execution was up to us.

In these situations, the consultant is brought in because she may look at the problem differently. People in the company may think, “it has to be done this way,” while the consultant may have different experiences, so she knows that’s not true. There are different and better ways out there; constraints aren’t binding how the company thinks, and it takes an outsider to recognize that.

So the consultant learns about the company and proposes ways of improving the systems. It may be that the consultant is completely wrong. The insiders at the company know lots of things that the consultant doesn’t, and it is impossible to convey all that local knowledge to the consultant. That is why in markets, it is ultimately up to them to make the decision.

The real advantage is that the consultant can propose things that the business never even considered. The company never imagined that this constraint wasn’t binding. The entrepreneur puts products in front of people that they never imagined they wanted. When Coase proposed spectrum auctions, he was putting forward a policy that people hadn’t even thought of. His unique approach to property rights economics allowed him to see the opportunity. Short of such extremes, the economist can propose smaller changes that amount to “you recognize the constraint exists, but it’s not quite how you imagine it.”

I like the analogy to consultants because it makes clear that the economist has no special scientific standing in the debate. The business consultant and the policy recommender never know whether they are correct. As I explained in a previous newsletter, we need to take seriously that our model of the world is wrong, just as the consultant might not know all the parts of the business. Maybe the constraint we think isn’t binding actually is binding. We need to consider that possibility.

Instead, it is better to recommendations as tentative hypotheses to test. In markets, the recommendation is subject to a market test. Do companies adopt it? Do the companies that adopt it make a profit and survive? James Buchanan argued that we should take the same approach in politics; recommendations are hypotheses to test. The test is less clear cut, but it’s better to have some test of our model than none. I don’t want to be a dictator simply declaring what is optimal.

I started thinking about this problem again after listening to this week’s EconTalk with Mike Munger on “The Perfect vs. the Good,” or what Mike calls directionalism vs. destinationism. A directionalist is hoping for incremental change in the right direction, while a destinationist will only accept jumping to the right destination. Mike argues that it is important to operate at both levels. I agree and would frame it as “it is important to think through what to do with various amounts and types of constraints.”

The art of making policy recommendations (which I struggle at but am working on) involves finding constraints that are not binding and persuading people and policymakers to take different actions given this realization. It is to ask: how can we move in the right direction on the margin?

---

Speaking of policy recommendations, I’m in The Telegraph ($) today arguing that the UK’s decision to block Microsoft’s acquisition of Activision/Blizzard. While Microsoft is far behind Sony on physical video game console market, it has pulled ahead in terms of “cloud gaming,” which doesn’t require an expensive Xbox or PlayStation. They are basically punished for being first into a new market.

If policymakers punish companies for creating new markets, they are punishing innovation, which is the last thing the UK needs. Instead, regulators should encourage companies to improve their products, even if we don’t know exactly what that will look like beforehand. Through all the chaos of this competition, consumers reap the rewards.