Questions for Interventionists

In an earlier post, I pointed out that we do not exist in a state of nature.  As many economists have been pointing out since at least Ronald Coase’s famous 1960 paper The Problem of Social Costs, we exist in a complex world of pre-existing social, economic, legal, and legislative arrangements.  These arrangements influence our actions.  Like Chesterton’s Fence, we cannot pretend they do not exist, nor discard them because we do not understand their purpose.  

And yet, many interventionists do ignore current arrangements.  The models they use ignore important aspects of reality, aspects that may show their interventions will do more harm than good.  Models are an important aspect of any reform proposal, but they are hardly sufficient.  Just because some model implies some desirable outcome does not mean the outcome will come about in reality.  Further, since models are dependent on their assumptions, any model can be used to justify any outcome.  

To prevent this issue of dueling models, allow me to propose some questions for interventionists.  These are questions to help justify their proposed interventions.  But first, two quick comments:

First, I am using “interventionist” is a very broad sense to mean any scheme that uses the power of government to intervene in economic relations for any purpose.  Interventionism includes (but is not limited to): market failure corrections, non-revenue taxes/subsidies (eg Pigouvian taxes), protectionism, industrial planning, nudges, etc.  

Second, I am placing the burden of proof squarely on the shoulders of interventionists.  In this sense, my approach here is very conservative: the status quo is taken as preferred over change unless shown otherwise.  This burden of proof can be met and overcome.  In that sense, my approach here promotes change.  The point here is to avoid radical and unproductive changes often advocated by interventionists while allowing potentially useful changes to arise.  

With the preliminaries out of the way, let’s get to the questions.

Question 1: What is the current state of affairs? 

This question is important because it sets the stage.  Of course, it is impossible to articulate every single aspect of the current state of affairs.  Rather, one should focus on the most salient (eg, direct laws, institutions, etc).  

Answering this question also helps prevent critical empirical errors that most interventionists make.  For example, if you ask almost any advocate of a carbon tax in the US, they will say “there is no price for carbon in America.”  That statement is factually incorrect.  There is no monetary price, sure.  But there is a price of carbon.  There are all sorts of preexisting arrangements that influence the price of carbon.  These preexisting arrangements, as Coase pointed out, are crucial.  If they are misunderstood, then interventions can make the situation worse.

Answering this question also helps understand why existing patterns are what they are.  And that leads us to our next question.

Question 2: Why have pre-existing arrangements failed?

If the answer to Question 1 leads one to conclude that there is indeed a failure, now we need to understand why that failure has occurred.  Is there something about the current state of affairs that triggers that failure?  What are the actual causes of the failure?  What are the incentives people face?

Understanding both pre-existing arrangements and why they fail help prevent cascading failure, where a mistake keeps getting repeated and repeated.  For example, a justification for tariffs is that trade can displace workers and it may take them time to adjust.  But current programs, like Trade Adjusted Assistance and Unemployment Insurance already exist to take care of those problems.  Research shows, however, that those programs actually extend the time it takes for workers to adjust to trade shocks.  Why have they failed?

In many cases, interventionists just assume the cause and go from there.  For example, it is often just assumed that public goods cannot be provided by the market in optimal quantities.  But research by Ronald Coase, Elinor Ostrom, and even Adam Smith shows that is not the case; public goods are often provided in sufficient quantities.  Despite the models, interventions could cause a failure to appear where there is not one.  

Question 3: Is your proposed solution the best method achievable? 

Hopefully, by this point, the interventionist has a pretty good understanding of the current state of affairs.  Now is the time to start considering proper interventions.  Note that this question actually has two elements to meet: 1) the intervention is the best method to achieve the goal, and 2) the intervention is achievable.

There are many ways to address this first element.  The word “best” here is deliberately vague and subjective: what is “best” will ultimately be determined by the analyst given their goals and preferences.  Consequently, there are many ways to determine “best.”  Calculating net present value is one way.  Utilitarianism is another way.  And so on. 

But what is best may not be a positive intervention (meaning that one takes a new action) at all.  Indeed, while investigating Questions 1 and 2, one may discover that the best thing to do is remove an existing intervention!  

The second element relates back to our first question.  Whether or not some intervention is achievable will depend on the current institutions.  A system where policy is decided by direct voting will have different achievable options than a dictatorship, which will have different options than one where a deliberative body acts, etc.  

Considering achievability will also force the interventionist to come to terms with the data they have.  We rarely have the data we want.  Costs and benefits are subjective and psychological; they depend on the situation one faces.  Monetary costs still matter, of course, but they are not the same as total costs once we move into collective decision making (for more on this point, see James Buchanan’s Introduction in L.S.E. Essays On Cost).  Confusing the two has led to many interpretive mistakes.  

These three questions are just the beginning.  Answering these can help shape interventions, but they still do not justify them.  More questions abound: ethical questions, political questions, legal questions.  But I hope they can provide a useful framework for discussing reforms.