Last week, Bob Solow passed away at 99. If I gave a list of influential economists to undergraduates and asked them to identify the Nobel laureates, I think Solow would be one of the easiest for them to guess (along with Coase). The Solow model of economic growth is so influential that every undergraduate economics student learns it and every growth model is benchmarked by it. His work is so influential, Obama gave him the Presidential Medal of Freedom.
I want to talk about an application of the Solow model, but I won't be discussing the model per se. The model gives a causal story behind economic growth, the kind that has policy insights. If you want a summary of some of the model's interesting insights, you can read this post by Noah Smith. I liked how he applied it to China's growth miracle.
But I want to look at how the model's growth accounting can be applied to development accounting.
Solow Model and Growth Accounting
The basics of the Solow model are pretty simple. The economy is a giant stew with three main ingredients: physical capital, human capital, and total factor productivity. Think of it as the meat, vegetables, and everything else that goes into the stew.
Economic growth happens when you get more stew. The Solow model says that to get more stew, you need to increase your ingredients. Want twice as much stew? Double the ingredients! But you don't have to increase all ingredients at the same rate. You can get more stew by adding a little bit more meat. Or maybe you add some more water and seasonings. The stew's taste won't be the same, but you'll get more stew nevertheless.1
Growth accounting is when we look at how we got more stew. We compare the contents of today's stew with yesterday's stew and see what changed. How much more meat is there? How much more vegetables? Is there more water? We account for the factors that we think are most important, then we say something like, "10% of stew growth came from adding more meat, 40% came from adding more vegetables, and the remaining 50% came from increasing everything else."
When we look at the economy, we ask ourselves how much the inputs to the economy grew. How many new machines did the economy get last year? How much education and work experience did the labor force gain? We're trying to understand how much growth came from physical capital and how much from physical capital. We won't capture everything, so we put any remaining growth into a residual. The Solow residual. Appropriate name.
If you want a good book on how we use growth accounting, I recommend Fully Grown by Dietrich Vollrath. He explains the details behind physical capital and human capital, and then he looks at growth patterns in the US. It's my top recommendation for people interested in learning more about macroeconomics.
Development Accounting
Growth accounting takes the same country and compares it to itself over time. What does the US look like today, what did it look like 10 years ago, and what explains the difference? But another way to apply the model is to compare different countries today. What does the US look like, what does Haiti look like, and what explains the difference? This is called development accounting.
I find this incredibly helpful because the most important question in economics is why are some countries rich and other countries poor. Development accounting at least gives us a place to start looking.
And what do the data say? The same Penn World Tables that allow us to do growth accounting also let us look at development accounting. In fact, the Penn World Tables reported their development accounting in this paper.
The weight to each factor is listed in each column. The different rows are checking the sensitivity to how we define physical capital. But they all say the same thing. Less than 10% of the difference between rich and poor countries is due to differences in physical capital. About 25-30% comes from human capital. And the remaining 65-70% comes from the Solow residual (TFP).
Now, let me issue my warning. As Brian Albrecht says, we should never reason from an accounting identity. We took the Solow model and turned it into an accounting identity to try and figure out why some countries are rich and some are poor. This does not mean that we can just increase human capital and expect countries to grow anymore than we can increase government spending and expect the economy to grow. (Indeed, I did a video showing how a strict focus on education produced some pretty terrible results.) We need to understand why human capital is lower, why productivity is lower.
But the accounting at least gives us a starting point. It tells us that when people say education is the key to lifting people out of poverty, there is some data that suggests that claim could be true and we should investigate it more. It tells us that when people start talking about the importance of strengthening property rights or reducing corruption, they're talking about ways to improve TFP.
I will always remember Solow for his contribution in how we understand developing countries.
Tech that Made Me a Better Economist in 2023
I normally make a video on my favorite tech of the year. This year, I got caught up with my trip to China, so I'll just describe my favorite tech to the paid subscribers.
Keep reading with a 7-day free trial
Subscribe to Market Power Newsletter to keep reading this post and get 7 days of free access to the full post archives.