Economic growth is the single most important indicator that is used in order to compare the developments of economic prosperity of different economies and in order to measure phases of economic boom or recession. In this context, the figure that one can normally read about in the news and that is most often presented in general, is the growth rate of real gross domestic product (gdp). It tries to measure the growth rate (usually on an annual basis) of the value of all final goods produced within a country, adjusted for price changes. Gdp growth is often praised as a solution to many economic and social issues like poverty, high levels of indebtedness and economic inequality. The fact that a recession is defined as at least two quarters in a row with negative real gdp growth and its definition hence relies solely on economic growth as an indicator underlines the important role that real gdp growth plays in the economic view of the world. In recent years, the slow and still ongoing recovery after the Great Recession in the developed world has spurred new discussions about long-term economic growth developments, the requirements for economic growth and its merits. But before diving into the topic of slowing economic growth rates, consider the possibility that the so-often-utilised real gdp growth is not a very reasonable indicator to measure developments in economic prosperity.
While the growth rate of total economic output might be a good measure for the development of the political power a country possesses, it is flawed when one wants to depict the development of the economic prosperity of a country’s inhabitants. This is due to one simple fact: Real gdp growth depends on both the growth rate of the population and of the economic output per capita. If one is interested in issues of economic prosperity and productivity of a society, it is therefore more reasonable to ignore population growth and solely look at the growth rate of real gdp per capita which measures the growth of the adjusted value of all final goods produced within a country divided by the number of inhabitants. Looking at the examples of the United States and Japan illustrates the difference this can make.
In the US, real gdp growth has consistently been positively influenced by the growth of the domestic population. While in the 60s, population growth accounted for about 1.3 percent of total gdp growth, this number declined steadily to about 0.75 percent in the recent five years. In Japan on the other hand, population growth has become negative in recent years and is dragging down real gdp growth with on average -0.17 percent over the past five years (in the 60s, population growth still contributed 1.23 percent to real gdp growth on average).
It follows that the narrative of Japan as the “sick man of Asia” is partly due to the fact that Japan moved into the age of a declining population earlier than other developed countries and hence has comparatively lower rates of real gdp growth than when one looks at gdp per capita growth. So far, during the second decade of the 21st century, Japan on average even had a slightly higher growth rate of per capita gdp than the US. Naturally, countries with lower population growth rates will face more periods that are defined as recessions than countries with the same gdp growth rate but a faster growing population. To avoid the demonstrated bias that population growth can introduce into real gdp growth when one is actually first and foremost interested in developments in economic prosperity, I will consider real gdp growth per capita henceforth.
In simplified terms, gdp per capita can grow because of three reasons. First, a larger part of the population can join the labour force. This happened to a large degree in most developed countries during the last century after the Second World War when more and more women started working while men’s participation rate only decreased slightly and longer schooling durations could not slow down the growth of the overall participation rate substantially. Second, the working population can work longer hours. In most developed countries, hours worked per worker have been consistently decreasing over the last decades (source: OECD). Third, the productivity of a worker, measured as economic output per hour worked can increase. An increase in productivity is the only way to grow economic output per capita in the long run (there is only a certain share of the population able to work and there are only 24 hours in a day).
How has economic growth been developing in the world’s biggest economies? While some economic developments are probably not as worrying as they might seem when only looking at total real gdp growth, one can observe a general slowdown of economic growth in the developed world when considering per capita data as well (I included China in the chart below for comparison).
In most developed countries, gdp per capita growth has been declining since the 1960s (if the substantial rise in the US, Germany and Japan in recent years is due to a long-term trend or is just caused by a rebound effect after the crisis remains to be seen). Recently, a lot of focus has been put on explaining the economic stagnation or slow recovery after the Great Recession in the developed world. Popular theories are a slow, ongoing reduction of debt overhang (supported, e.g., by Harvard professor Kenneth Rogoff, former chief economist at the IMF), secular stagnation (a term coined by Alvin Hansen in 1938) due to demand side mechanisms (with Larry Summers, another Harvard professor and former Secretary of the Treasury, being the most famous proponent) and secular stagnation due to supply side mechanisms (see Gordon, 2015). However, while all of these concepts present a good explanation for at least parts of the slow recovery in the aftermath of the financial crisis, not all of them can explain why economic growth has been slowing for about half a century. Deleveraging as described by Lo & Rogoff (2015) has only become an issue within the past decade, as debt volumes have been growing consistently beforehand. Hence, although all three theories may possess some explanatory power for the recent sluggish economy and should be considered when devising new policy approaches, in order to understand the long-term decline of economic growth, looking at both the demand side as well as the supply side story appears to be the most promising.
Regarding the demand side, Summers (2014) argues that the recent period of slow growth is caused by a “full-employment real interest rate” (a definition rather similar to the traditional natural interest rate) that is too low to be feasible with conventional monetary policy and hence prevents savings from matching investments on a level that enables full employment. In this scenario, something has got to give and, consequently, gdp has to fall below its potential until investments match savings again. Although it is very hard to believe that investments could not match savings due to the constraint of the zero lower bound already back in the 70s and despite the fact that economies like Germany and the US are arguably quite close to full employment at the moment, it is still worth considering some of the underlying causes on the demand side for low growth rates that Summers mentions. First, demand for capital goods might have declined due to the change in composition of modern economies as information technology companies operate less capital-intensively than more traditional ones. Second, rising income and wealth inequality raises the share of income going to individuals with a lower propensity for consumption and a higher propensity for saving. This lowers consumption and puts downward pressure on interest rates due to higher saving demands and, if rates cannot adjust in a way that allows investments to match the high savings, gdp will decline. Third, increasing frictions in financial intermediation and a run to save havens (e.g., government bonds and central bank deposits) can work to raise the wedge between safe liquid rates and rates charged to borrowers. Additionally, demographic changes can also lead to a higher demand for saving as old workers have a higher propensity for saving than young workers who tend to rely on their human capital. All these factors can lead the level of output to be below its potential due to the lack of investments and a too high saving rate. This mechanism can even have a long-term impact as lower levels of investment can stunt future growth.
However, in the long run, economic growth is mainly driven by supply side mechanisms (i.e. the growth of potential output). Gordon (2015) names two main reasons for lower growth paths of per capita gdp. First, hours worked per capita start to decrease due to the retirement of the baby boom generation after steady increases because of a growing participation rate among women until the end of the last century. And second, and perhaps more importantly, productivity growth is slowing. This could have various reasons. Educational attainment is already high and may not be easily improved further, especially if the share of children growing up in low-income households will increase. Inventions and innovations in information technology do not seem to have the same growth-spurring effects as older ones had (e.g., a self-driving car may not be the same to a regular car as a regular car was to a horse). Finally, the effects of increasing automation are ambiguous: While robots work faster and cheaper than human workers can, they also aid the increase of economic inequality (due to a further decreasing labour share of national income; see OECD, 2015) which in turn might stunt growth. Summarising, the supply side narrative states that the low-hanging fruits have already been picked and further economic growth will be slower.
Two things are worth noting at this point. First, there are probably no easy fixes to the problem of declining growth rates as the supply side story of sinking output potential appears to be able to explain the long-term slowdown of economic growth fairly well and, additionally, most of the underlying causes that might trigger a demand side stagnation are also of a long-term nature (IT, inequality, demography). Instead of short-term fiscal or monetary policy schemes that bring a shocked economy back into balance, long-term investments in education and infrastructure and mitigation of economic inequality in order to manage the effects from shrinking and aging populations seem to be more promising. Second, it might be time to accept the fact that the times of sustainable growth rates above two percent are over. Strong economic growth is not a rule of nature. A look at the database provided by The Maddison-Project shows that before the 17th century, most countries grew at an average level close to zero percent. Historically, the post Second World War growth has been a remarkable exception, not the rule.
The outstandingly high growth rates in Germany and Japan in the 50s and 60s are mostly due to the fact that they had a lot of room to catch up to other developed countries after having been bombed on a never-before-seen scale for multiple years (Japan’s approximated gdp per capita growth in 1945: -49.3 percent). China has only been able to outgrow the Western countries in the past couple of decades because it can adopt technology that has already been used in those countries for years (On a per-capita level, China’s economy is far from overtaking the developed world; it is merely reducing the still huge gap). Nobody knows if new inventions will lead to a new economic miracle in the near future. However, it might not be advisable to reckon with it. Developed countries should try to implement policies that take into account the possibility that they will stay on more conservative growth paths with still positive, but lower productivity growth.
I will look at implications of slowing growth rates as a secular rather than a temporary process for monetary policy in a future article.
Additionally, I will publish an article in the not-too-distant future that discusses the concept of gdp and whether it is a good measure of economic prosperity or even general welfare and human development.
Gordon, R. J. (2015). Secular stagnation: A supply-side view. The American Economic Review, 105(5), 54-59.
Lo, S. H., & Rogoff, K. (2015). Secular stagnation, debt overhang and other rationales for sluggish growth, six years on.
OECD. (2015). The Labour Share in G20 Economies. Report prepared for the G20 Employment Working Group Antalya, Turkey, 26-27 February 2015.
Summers, L. H. (2014). Reflections on the ‘New Secular Stagnation Hypothesis’. Secular stagnation: Facts, causes and cures, 27-40.
The Maddison-Project, http://www.ggdc.net/maddison/maddison-project/home.htm, 2013 version.
written by Jonas