[MUSIC] How can we think more systematically about categorizing the sources of economic growth in China? Well probably we can considered three different sources of growth. The first is increasing inputs. So, if we put more in, we'll get more out. If we invest more capital and labor into an activity, we should expect to get more output. Second, we can increase efficiency. And here we can think of two types of efficiency. The first is allocational efficiency, allocating the available resources in the economy to the most productive activities. So obviously if I take resources away from a low return project or a low return firm and invest those same resources into a higher return project, or a more productive firm, we're going to get greater overall output. And similarly, we can think of reallocating resources from lower return sectors of the economy, like agriculture, to higher return sectors of the economy, such as industry. And this reallocation of resources across broad sectors of the economy is often referred to as structural change. The second type of efficiency is technical efficiency. Getting more output from the same amount of inputs. For instance, if workers for whatever reason lack incentives to work hard, they may really underperform and produce much less output than is possible, given the existing technologies. And finally, we think of increasing productivity. If we have new technologies, innovation, we're going to get higher output from the same amount of input. Now, how can we think about this more formally? Think about an economy where first output is at point y1, and then in the second period it increases to y2. Well in this figure we're plotting the output as a function of the amount of inputs X. And we're also allowing for the technology to change from Production Frontier 1 to Production Frontier 2. In this framework we can think about the three sources of economic growth in the following way. First, efficiency will be the change in the distance between the observed output and the production possibility frontier. So E1 is how much Inefficiency there is because the actual output is below what is possible. And if there is a reduction in that inefficiency, E1 shrinks to a level E2 which is smaller, than any difference is going to be to growth from increased efficiency. Now the second source of growth is just increasing inputs from X1 to X2, given that we're at the production possibility frontier levels. And the third source of growth is the amount of extra output we get by shifting to a new production frontier, again, at the efficient level of production. And so together, these three components add up to the total change from Y1 to Y2. Now how can we bring this back to what has happened in China during the period of economic reform? Well in terms of increasing efficiency, obviously one of the big factors to improve allocational efficiency has been the introduction of markets. Which gives producers the right price signals to produce things that consumers really value and care about. In addition, there have been a number of ways in which incentives have been improved to also increase efficiency. So, decentralization of decision making authority to farmers through the house of responsibility system, or to firm managers, through the enterprise responsibility system. Or even to local government officials to give them a stronger incentive to help firms and households produce more economic output, all contribute to greater technical efficiency. At the broader, macro level there's also been a structural change of movement of labor out of agriculture into industry, often through migration of rural laborers into Chinese cities working in factories. But also over time there's been an upgrading of the industrial sector to higher and higher productivity sectors. If we think about increasing inputs, I think it's worth pointing out that under the socialist planning period before reforms, this was really the only source of growth. Because China relied on mobilizing resources in the economy. Much of it by taxing individuals, taxing rural people in particular, and making big investments in heavy industry to expand output, but not necessarily accompanied by increases in standards of living. In the reform period there's also been steady increases in inputs. China, we know, has a pretty high investment rate. And that is a pretty direct result of having a very high savings rate. And so this has also contributed to growth. And finally, if we think about increasing productivity, improving the technologies that are being used in the economy, initially, by globalizing the economy. China was able to really transfer technologies from the rest of the world through foreign direct investment, through engaging in international trade and learning about the needs of customers globally. And over time, has started to shift more and more to an innovation system where Chinese firms themselves are becoming the sources of invention and new technologies. Now, if we think about how to quantify these different sources of growth quantitatively, we can think of a production function. Which relates the amount of GDP to a number of different factors, including a total factor productivity, term A, a capital input. And here we're also going to include the parameter alpha which kind of captures the relative importance of capital in total output compared to labor. We'll also include the number of workers, the other main input in economic production. And h which is a measure of the human capital of labor. So as the education level of workers improves, we should expect them to be more productive. Now starting with that formula, we can actually derive a decomposition formula which relates the growth rate of per capita GDP to four components. The growth rate of labor participation rate, the growth rate of the capital output ratio, the growth rate of the average level of human capital and the growth rate of total factor productivity. And when researchers have taken this formula to the data, what have they found? Well first, let me emphasize what total factor productivity means. If two companies have the same amount of employees and machines but produce really different levels of output, then that difference is really a difference in the total factor productivity. So total factor productivity is simply the portion of the output that is not explained by the amount of inputs used in production. And this could be due to differences in efficiency or differences in technology. Now if we look at how these different factors contribute to overall growth in GDP per capita, and here in the pre-reform, pre 1952 to 1978, the average growth rate was 3%. In the reform period, it was 8.1% on average. But if we kind of decompose the 100% of the growth in these two periods, which factors account for them, well the results are really striking because there's such huge differences between the two periods. The labor force participation rate really hasn't contributed very much in either period because most people in China are working. But the capital output ratio was the most important driver in the socialist period. Accounting for more than 100% of total growth, but relatively unimportant in the Chinese context. In the Chinese context by far the most important driver has been total factor productivity growth which has accounted for 78% of overall growth. And factor productivity growth has actually been negative during the socialist period. And then Human Capital has also been important in both periods. But relatively more important in the socialist period. So, what have we learned? In conclusion, it's important to emphasize that, during the socialist period, growth relied primarily upon increasing inputs. And, during the reform period, growth has come mainly from productivity growth. And human capital has also have played a role. There's still remaining questions of course. For example, how important has structural change been in explaining that total productivity growth which really didn't divide the productivity into the different sectors of the economy. And also what is the scope for improving efficiency in the future? Can we get more growth from that source as well?
