Degrowth: Delinking GDP, output and value

This is my contribution to the ongoing polemic between the proponents of degrowth and those who believe that degrowthers are engaged in mystical thinking ignoring the hard economic reality and the basic tenets of economic theory. The degrowthers advance a number of arguments accusing the opponents of empirical and analytical flaws. The opposing stances are well represented by Branko Milanovic who says that degrowth is bound to result in perpetual poverty in developing countries and reduced standards of living in the West (https://braveneweurope.com/branko-milanovic-degrowth-solving-the-impasse-by-magical-thinking) and Jason Hickel who defends slower economic growth (https://www.jasonhickel.org/blog/2017/11/19/why-branko-milanovic-is-wrong-about-de-growth).

My own interest in the subject (apart from the fact that the ongoing discussion is indeed very interesting and intellectually stimulating) was kindled by my recent work on modelling local economic development in the context of UNCDF (UN Capital Development Fund) programming. Even earlier, when I worked on the diagnostic tool for urban economic resilience, the question of whether GDP growth should be considered as an indicator of development that contributes, among other things, to economic resilience or not, loomed large. In both cases I’ve resolved the question by accepting per capita GDP as one of the major outcome indicators.

The strongest argument advanced by degrowthers is that GDP is simply not a right measure of development and when it comes to human well-being, counting GDP is irrelevant. This is, of course, nothing new in mainstream economicsand a number of prominent representatives of the profession including Stiglitz and Mazzucato among others, criticized the inherent flaws of the GDP methodology. Incidentally, Simon Kuznets, the creator of GDP, was well aware of these flaws, famously saying that “Distinctions must be kept in mind between quantity and quality of growth, between its costs and return, and between the short and the long term. Goals for more growth should specify more growth of what and for what.”  

Milanovic also recognizes the imperfections of GDP but I agree with his opinion that while it is imperfect at the edges, it’s fairly accurate overall as an indicator of development. To argue otherwise, would mean to ignore the historic reality. At the same time, the level of GDP is not the same as the GDP growth rate. Starting from a low base and far from the production frontier allows a country to grow faster than countries at higher levels of development of closer to their production frontier. We all know that many developing countries grow much faster than developed economies (in 2010-2020, the US average annual growth rate was about 2% and that of Uganda about 6%, to take just one example). After all, the unconditional convergence theory postulates that all countries will eventually achieve the same rates of growth. Also, GDP growth rates in different countries have dramatically different consequences on the resources use in those countries, not to mention the overall impact on global environment and climate change.

Where I don’t quite agree with Milanovic’s argument and am sympathetic to the degrowers’ is that GDP is equivalent to the quality and quantity of goods and services available to the population. Leaving aside all the controversies around GDP methodology (it leaves out non-commercialized activities that are welfare-enhancing) and assuming that the same methodology continues to be used, I see an opportunity for delinking GDP from the quality and quantity of output.

The biggest strength but also the biggest weakness of GDP is that it relies on the monetary value of final goods and services produced during a year calculated as their market prices. We don’t need to go into the vagaries of price formation under market conditions. As is well known, only efficient markets clear the prices at the marginal production costs but – alas – efficient markets do not exist. We know that markets can be irrational and “exuberant” (to borrow from Robert Shiller). If we delink the value of GDP from the value of goods and services, it becomes possible to increase output without a commensurate increase in GDP or increase GDP without an actual increase in output (which is the major concern of the degrowthers because of its impact on resources and the environment). And I don’t mean green growth, to which Hickel objects as “more growth entails more energy use, and more energy use makes it all the more difficult to cover that demand with renewables”.

Output growth overtakes the GDP growth. As discussed, both quality and quantity of output matter. It’s a well-known fact that new products, particularly those technology-based, offer better user experiences and open up new possibilities although their monetary value may not differ much or at all. The price of cheaper computers is now comparable to the price of typewriters some 40 years ago although the difference between the two products is obvious. Furthermore, as new products enter mass production and technologies refine, the productive efficiency increases resulting in a low price per unit. So, whereas the output in terms of material product accounting (which used to be the favorite in the former socialist countries) grows, GDP may stay at the same level (or even decrease). I’m leaving aside the question which production is more resource consuming and resource intensive – typewriters or computers.

A well-known example is structural transition in developing countries with an increasing share of higher-productivity economic sectors that produce greater quantities (and hopefully, better) goods and services without the matching increase in the overall monetary value of the output. But the trend is observable even in developed countries in the context of the gig economy. Milanovic (Capitalism, Alone https://deveconhub.com/inequality-and-capitalism-alone/) highlights this trend in connection with the growth of the gig economy that commercializes our free time and things that we know but have not used for commercial purposes before. Uber, for example, can allow paying much less to its drivers because they are working part-time and using their own assets whereas the quantity of the output (number of trips in this case) may increase. At the same time, the GDP is likely to remain the same or even decrease because the Uber business model does not envisage any capital investments (which have already been made by individuals in the form of their cars, maintenance facilities, etc.). Again, it is highly desirable that such sectors are more resource- and energy-conserving but this is not a necessary condition for the output to grow without affecting GDP.   

GDP growth overtakes the output growth. But the other interesting implication is that it is possible for the GDP to grow without any increase in output. One reason for that (which Milanovic explores in Capitalism, Alone) is the commodification of activities that used to be conducted within families (and not part of the GDP calculation) are now outsourced and delivered by formal service providers, e.g. food cooking which used to be a purely family output or dog-walking. It has two implications. It is important to not however, that what we have in this case is an accounting transfer of some activities that have not been previously accounted for without the actual increase in the output (or resources required for production).  

The other aspects are related to the methodological issue around the GDP, for example, how value added is accounted for in GDP and market price fluctuations. This concerns, first of all, the financial sector. The 1993 SNA (System of National Accounts) began the process of counting FISIM (financial intermediation services, indirectly measured) as the financial sector’s value added to the economy. Introduction of this approach (FISIM is calculated by the extent to which banks can mark up their customers’ borrowing rates over the lowest available rate) has resulted in a jump in the financial sector contribution to GDP in the US and UK from zero a few years before to 7.3% and 7.2% respectively in 2016. Apart from the fact that the validity of FISIM as a value addition measure is questioned by many (see for example, Mazzucato’s The Value of Everything, https://deveconhub.com/mariana-mazzucato-the-value-of-everything-part-1/), it’s obvious that this leap in GDP had nothing to do with the actual output. If, for example, Indonesia introduces FISIM in the near future (as it is planning to do), its GDP will increase but the actual production will not.

Another methodological issue is linked to GDP rebasing. While nominal GDP measures the value of products and services (including the price and volume), the constant price GDP measures any changes in the volume of production, eliminating the changes in the prices of products and services by keeping the price level constant at base year levels. Thus, real GDP depicts a clear picture of changes in the actual production level of the country. Rebasing of GDP means replacing the old base year used for compiling the GDP with a new, more recent, base year for computing the constant price estimates. This is how Nigeria became overnight the biggest African economy overtaking the South Africa. In 2013, Nigeria’s GDP was rebased from about USD 270 billion to USD 510 billion, a 90% leap. But selecting a new base year (with higher prices) for GDP calculation is not related to an actual increase in output.  

The other possible reason for changes in the GDP without changes in the output is economic. Price fluctuations, especially for commodities, may cause significant changes in GDP and conceivably as some natural resources become scarcer, their prices may increase long-term contributing to higher GDP at the same or lower level of production. As the prices of natural gas have jumped 40% in Europe in October, Russia is likely to show a greater GDP for 2020 (if its existing contracts for gas export allow price increases) even if the level of production remains the same.

Lastly, the recent demographic trend in some developed countries indicates the possibility of an increase in GDP/output per capita without any changes in output or the value of final products. In normal circumstances, as the population increases, economic growth is driven by demographics and a purely economic component, as Thomas Piketty emphasized in Capital in the 21^st Century.  GDP can be decomposed into its population and economic elements by writing it as population times per capita GDP. Expressed as percentage changes, economic growth is equal to population growth plus growth in per capita GDP. But the pronounced demographic trend, especially in South-East Europe, has been decline in population and depopulation. For example, Serbia’s population dropped from 7.5 million to below 7 million in the past 20 years. According to the World Bank, Serbia’s population is projected to fall further to 5.8 million by 2050. That would represent a 25% fall since 1990.This trend is mirrored in some other countries in the region: Bulgaria, Bosnia and Herzegovina, Albania, Northern Macedonia.

Still, even in this situation the GDP growth is not entirely impossible due to the factors that have been discussed above. But it is more likely that the output (and GDP that measure it) either stagnate (if the drop in the population share is offset by increased productivity) or decline. A stagnating GDP against the backdrop of a declining population indicates, by definition, higher GDP per capita. If GDP is decreasing, then this is a matter of the respective rates of change: if population is decreasing faster than GDP, GDP per capita will continue to increase.  This scenario, limited to a handful of countries at the moment, may become more common as many developing countries (including those in Africa) are completing their demographic transition.

I intentionally do not consider the environmental impacts of GDP (or output). It’s conceivable that even the lower levels of GDP and/or production could be more energy-consuming or more harmful to environment.  However, the present trend in technological development and the growing public awareness about the detrimental impacts of climate change (coupled with a more pronounced political will) give rise to hopes that the future growth, whether in terms of physical output or as an accounting outcome, will be more environmentally friendly. However, it is important, from the methodological perspective, to distinguish between the output in terms of goods and services on the one hand and the GDP calculations on the other.