In 1815, after Duke of Wellington ends years of war between the United Kingdom and France by defeating Napoleon at Waterloo, the UK parliament passes the “Corn Laws” which prohibit import of cheap grain from France. The motivation is obviously to protect British land owner’s monopoly, but the argument for why the UK shouldn’t be buying French grain is based on Adam Smith’s principle of absolute advantage. British farmers can produce more grain than the same number of French farmers, so why would the UK be buying grain from France?
Portrait of David Ricardo by Thomas Phillips
To refute this argument, a British economist David Ricardo described what is perhaps the most complex and counterintuitive principle in economics. It is called the theory of comparative advantage, and it feels more like a natural law than an economics principle. It is definitely worth studying in depth, but here we only have space for a brief example.
Imagine two castaways who must survive by fishing and gathering berries. One of them is competent, and it takes her 4 hours to catch a fish, and another 4 hours to gather 100 berries. The other castaway is anything but competent, and it takes him 6 hours to catch a fish, and another 6 hours to gather 60 berries. The competent castaway has an absolute advantage over the incompetent one in both fishing and berry gathering, and without considering opportunity cost it seems that she has nothing to gain from trading with him. But let’s see what happens if they specialize, and focus on what they each have a comparative advantage in. If she doesn’t have to fish, the competent castaway can gather 200 berries in her 8 hour workday. The incompetent one can catch 2 fish in his 12 hour workday. He will gladly trade one fish for 75 berries because he knows that he could gather only 60 berries in the time it took him to catch that fish, and she will gladly trade 75 berries for one fish because she knows that it would take her an hour longer to catch a fish than to gather 75 berries. They still work the same hours, each one still eats one fish per day, but he eats 15 more berries, and she eats 25 more berries each day. That’s the magic of comparative advantage. The extra 40 berries worth of value are created simply by specialization and trade.
So how is the theory of comparative advantage an argument for cloud computing?
There are many benefits of cloud computing, but many IT departments resist it for the same reasons British farmers resisted importing cheap grain from France 200 years ago. Most of the benefits of cloud computing are obvious and difficult to argue against, so the IT departments’ main argument against adopting cloud computing is that they can deliver the same benefits better and/or cheaper. Such claims are usually not true, but they are difficult to disprove at the time the decision is being made. So let’s give the IT departments the benefit of the doubt, and assume that they can indeed provide all the benefits of cloud computing better and/or cheaper than Amazon, Google, Microsoft or any other public cloud computing provider. The theory of comparative advantage says that, even in that case, most companies would be better off adopting cloud computing as much as possible, and focusing on their business instead of worrying about IT. Presumably, these companies have comparative advantage in whatever their business is, while the cloud computing providers have comparative advantage in providing IT as a service.
It is certainly an interesting argument, and has a degree of truth. I think it would be most true for smaller companies, where focus on their comparative advantage would be good. For larger companies, the IT department is completely separate and the only likely result would be shrinking that department. The overall enterprise would not focus more on their business. Therefore for larger companies, I think the benefits of cloud computing have to stand on their own, but for smaller companies, this argument could be persuasive.
As owner of a tiny company, I know that it is downright insane for me to do any of my own hosting or administration, aside from what I need to do in R&D. I sometimes do it anyway, but it makes no sense.
Thank you for the comment Ben. You are right that large IT departments can act as separate entities from the rest of their enterprises. To realize the benefits of comparative advantage in that case, they must provide IT as a Service (ITaaS), which means they must offer On-Demand Self Service, Meaningful Service Level Agreements, Service Metering, Resource Pooling, Rapid Elasticity, and so on. Many IT departments have tried to implement ITaaS model using virtualization and even cloud computing, but there are two problems with this approach. The first problem is that most are not large enough, and simply lack the scale necessary to do this successfully. The more interesting problem is that when they succeed in implementing ITaaS, they discover that there is not much difference between providing ITaaS to consumers within their own enterprise and providing it to anyone else. Some in that situation may choose to become public cloud computing providers, but even if they continue to serve IT service consumers within their own enterprises exclusively, the theory of comparative advantage remains an argument for cloud computing. This does not mean that there are no arguments against cloud computing, or that adopting cloud computing is always the right choice. (Even the competent castaway may prefer solitude or enjoy fishing, and choose to forgo the extra 25 berries per day for those reasons.)