Coase was a very interesting and hugely influential scholar. His two most cited works, and they are really heavily cited, use the idea of transaction costs and externalities to ask, why do firms exists in the first place? And how much does the law (specifically, property rights/liability assignation) matter when it comes to determining efficient outcomes? The latter gives rise to the unfortunately named Coase theorem*, which is often misstated as saying something like “it doesn’t matter who gets assigned the property rights, the efficient outcome will always happen.” The classic example is a farmer and a shepherd on adjacent plots of land. The question is, should they build a fence? Well, if the sheep do damage to the crops greater than the cost of building a fence, it’s efficient to build the fence. What Coase argues is that, in the absence of transaction costs, the fence will get built if it is efficient, and will not get built if it is not efficient, no matter who has the liability for the damages – that is, no matter the shepherd is obligated to pay for the damages or not. The farmer and the shepherd will come to an appropriate side deal (“Coasian bargaining”). What often goes unsaid, however, is that the assignment of property rights does affect the distribution – who ends up paying the most. The paper can also be rephrased to say that since transaction costs do exist, changing the assignment of liability might produce a more or less efficient outcome.

All that is to say that Coase is both very influential and a bit different than your typical economist. In 2002, Coase gave a lecture where he argued for greater realism in economics. Coase argued that economics had not changed sufficiently since the days of Samuelson, Marshall, or even Smith. Coase asserts that economics will change, and that it ought to, and in part that it must change by expanding its scope. But rather than embracing the economic imperialism of Gary Becker, Coase argues for the opposite:

Gary Becker talks about an economic approach. In effect what they are saying is that economics is a bag of tools, a way of analyzing problems, and it no doubt is. It has resulted in what is termed “economic imperialism,” namely, taking those tools and analyzing other subjects. … All this explains, I think, why economists are so happy. They have these tools, they are useful, they go around not only improving economics but do a lot to improve other subjects.
…
Now it is true that there have been a lot of complaints about formalism in economics – the elegant but sterile reasoning and so on – but also talks about the lack of realism in economic discussion.
…
What I think is important is that economists don’t study the working of the economic system. That is to say, they don’t think they’re studying any system with all its interrelationships. It is as if a biologist studied the circulation of the blood without the body. It is a pretty gory thought, but it wouldn’t get you anywhere. You wouldn’t be able to discuss the circulation of the blood in a sensible way. And that’s what happens in economics.
…
Transaction costs, in my view, become the factor upon which the productivity of the economic system depends.

However, transaction costs depend, as we learned from the new institutional economics, on the working of the legal system (the system of property rights, the enforcement of property rights, the ability to foresee what the legal decisions will be, and so on). They also depend on the political system, they depend on the educational system, and they are interrelated with other social systems. And in consequence, economists should enlist the support of lawyers, sociologists, anthropologists, and others in our work in order to understand why transaction costs are what they actually are. It’s the opposite of economic imperialism. We should invite these other practitioners in these other fields into our realm to help us in understanding how the economic system actually operates.

Ok, I excerpted a few big chunks, but I like a lot of the moves Coase is making. The rest of the speech (which is quite short) is also worth your time, I think. Coase makes fun of formalism from the “left” (Keynes, Hicks) and “right” (Becker) because they all treat economics as a way of thinking rather than an empirical exercise in understanding how a system works. Though I don’t know if I’d be so quick as to lump Smith in with Marshall and Samuelson…

*Unfortunately named because it’s not really a theorem, and when people think of it as such they often misremember it. I have tested this on a small sample of law students (who learn the Coase theorem in their contracts or property class). See also Misinterpreting the Coase Theorem.