The usefulness of the GDP measure

We hear a lot about GDP - but what exactly is it and should we pay much attention to it?

Summary

• Gross Domestic Product (“GDP”) is a common measure reported in the media and used by economists

• It is a simple measure that values the overall economic output of a country

• A lot of importance is placed on the measure, although this importance is probably not justified

• People actually ‘feel’ different depending on the published GDP numbers

• GDP is not a measure of welfare, although it is often used in that context

• Because GDP is so important in our political discourse, it may lead us to focus our efforts on factors that do not actually improve welfare

• Numerous agencies and governments are working on new measures that control for quality of life metrics such as leisure and life expectancy. This is called “Beyond GDP”

Introduction

Last week, we got a slew of GDP data: US GDP grew 2.9% in the fourth quarter of 2022, German GDP went down by 0.2%, French GDP went up by 0.1%, Polish GDP went up by 2.2%. So what is GDP all about? 

One of the first things we are taught in any introductory macroeconomics course is the idea of GDP (Gross Domestic Product). GDP is also mentioned in many political contexts, especially focusing around economic growth. Recessions, which have a significant impact on the overall national sentiment, are usually defined through GDP (two consecutive quarters of a reduction in GDP is generally considered a recession¹). GDP is also informally used to rank the economies of countries, either in absolute terms (total GDP) or in per-capita terms (divided by the population size). Where did this measure come from, how useful is it, and how much attention should we pay to it?

Creation of “GDP”

The GDP measure was invented by Simon Kuznets, a Nobel-Prize winning economist around 1937. It started to be used globally in 1944, after the Bretton Woods conference.² The measure itself is relatively simple: it attempts to value the overall economic output of a given country by using the transactional value (in currency) of goods and services. Broadly, GDP (“Y”) comprises the value of consumption (i.e. goods and services you buy), investment (i.e. people's savings that are used for investing into things such as infrastructure, research etc.)³, government spending, and lastly net exports (i.e. the difference between exports and imports). GDP is simply an accounting identity, often written out as:

Y = C + I + G + NX

Over time, the GDP measure became an important metric in public discourse due to the strong positive correlation between the GDP measure and many other welfare metrics, such as life expectancy, nutrition, and infant mortality. However, it is worth noting that Kuznets himself did not want GDP to be linked with the broad concept of welfare, which can be thought of as the well-being of society.

Beyond the correlations GDP exhibits with important well-being metrics, the measuring and reporting of GDP has two additional important effects:

1. GDP figures, which in many countries are reported quarterly, influence monetary policy in significant ways;

2. Quarterly GDP has significant impacts on the perception of what is occurring in the economy.

1. GDP and Monetary Policy

Measures of GDP are important for Central Banks, especially for the US Federal Reserve. The reason why the US Federal Reserve is unique to most other central banks is that they have a dual mandate – to maintain price stability (i.e. control inflation) and to maintain full employment.⁴ The first goal of price stability is easy to monitor, as we have reliable inflation data and we have the tools to control inflation. Achieving the second goal is less obvious. This is because of the issue of what exactly does “full employment” mean. The concept underlying full employment is “Potential Output” or “Potential GDP”. Potential Output refers to the highest level of output that can be sustained over the long-term. This amount of output is restricted by physical constraints (such as available land, labor, and capital), technology level, and laws. An economy that has actual GDP below potential GDP is an economy with ‘slack’ in it. It could produce more, but some frictions (for example, too high interest rates preventing investment or inability of companies and individuals to access the capital markets) in the economy are preventing it from producing more. An economy that is above potential GDP is ‘overheating’. When the economy overheats, inflation tends to rise as the demand for the production factors such as capital and labor is higher than the supply. This is when the Central Bank tends to step in to ‘cool’ the economy by increasing interest rates, reducing demand for capital (as it is now more expensive) and increasing the capital supply (as you can now earn a higher return on saving, so you spend less).

Potential GDP is closely linked with the concept of the natural rate of unemployment. Natural rate of unemployment is a concept developed by Milton Friedman and Edmund Phelps, who both received the Nobel Prize in Economics for work on this concept. The natural rate of unemployment is the rate of unemployment that exists when the economy is at the long run growth level, or potential GDP. It is intuitively explained as a necessary rate of unemployment in order for the economy to develop successfully. The reason for this is to mitigate labor frictions. If someone would want to create a new company, having access to available labor makes the process easier. Likewise if certain companies would want to react quickly to a change in demand, having available labor makes it possible for them to hire rapidly. Of course, having a buffer on unemployed labor comes at a cost – the unemployed labor needs to receive unemployment benefits and it also adversely affects individuals' long run incomes. Thus, the question of what is the natural rate of unemployment is a difficult one, both from an economic perspective (as overshooting or undershooting it has significant cost) and also a political one.

2. GDP, Sentiment and Welfare

GDP numbers stir up a lot of discussion and economic commentary, especially when GDP growth is negative. In 2022, the US experienced two quarters of negative GDP growth. Typically, this would be called a recession, which is a general decline in business activity. Recessions in the US are announced by the National Bureau of Economic Research (NBER). However, the NBER did not call it a recession, because the underlying performance of the economy was better than the GDP numbers suggested. The reason 2 quarters of falling GDP occurred is because of how GDP is measured. The reason the economy contracted was broadly due to inventory mismanagement – companies over-ordered stock in the last quarter of 2021 and then did not need to make additional purchases throughout the first half of 2022, because of the backlog. The drop in inventory orders reduced GDP growth by 2 percentage points in the second quarter of 2022. GDP grew by 6.9% in the last quarter of 2021, but fell in Q1 2022 by 1.6% and Q2 2022 by 0.9%. If the companies had managed their inventories better, we would have seen a less volatile GDP growth rate.

GDP numbers, however, matter in influencing people’s perception of the economy and well-being. Firms and individuals make decisions based on whether the economy is in a recession or not. De Neve et al (2018) found that self-reported life satisfaction reacts more to negative GDP growth than to positive GDP growth. To offset the impact of a 1% GDP contraction on well-being, the economy would need to grow between 2% to 6%. Given the impacts on our well-being the GDP measure has, is GDP actually a good measure of welfare?

GDP, as can be seen from the definition, does not take into account many important factors that influence welfare. For example, GDP does not put any value on the negative (and positive) externalities arising from production, such as pollution. GDP also does not take into account the amount of remaining natural resources. Thus, if a country extracts more of its natural resources in a given year, its GDP will go up, but it might not be able to extract the same amount in the next year (this is similar to what happened with the inventory issue mentioned above).

GDP also does not give any value to notions such as free time. Reading a book to your child generates no GDP, and it might even reduce GDP if it is replacing child care for which a person would have to pay. This also creates a paradox – any new technology that saves us time by making us more efficient, may end up reducing GDP. because we might work less. Lastly, total GDP or GDP per capita has nothing to say about inequality and the unequal distribution of resources in the society.

GDP and Inequality

US real GDP per capita has nearly doubled in the time period 1976-2018 (total real GDP quadrupled in that time, as the population nearly doubled). How much of that growth, translated into income growth, has been seen by the bottom 50% on the income ladder? They have only seen an overall real income growth of 19%. That is an annualized growth rate of about 0.3%. This information is very well presented by Blanchet, Saez and Zucman, who have developed a model for a real-time income tracker. This means that for 50% of Americans, their incomes, from their perspective, have been flat. So, if you are in the bottom 50% of the income distribution in the US, does the headline GDP growth number matter to you?

This also leads to a re-thinking of the concept of a recession. As mentioned, a recession is usually announced after two consecutive quarters of negative GDP growth. From an individual’s perspective, even when the economy is growing, they may have experienced real income declines – for example, in the real income inequality time tracker, there are multiple time periods where income for the bottom 50% is falling, such as during the 2000s and 2012 as can be seen in the graph below.

From the perspective of these individuals, they are in a recession, as they saw real incomes (their own “personal GDP”) fall, while real GDP grew by over 40%, as growth did not impact everyone equally. Conversely, today we may have the opposite movements – as everyone is discussing a possible recession and real GDP falling – real incomes of the bottom 50% have grown by 3.5%, as can be seen in the graph below.

Thus, does it make sense to look solely at GDP when making judgements about the economy? It appears that for developed countries such as the US, it is not very helpful. Incomes of people in different parts of the income distribution respond very differently to overall economic performance, and focusing solely on GDP may result in erroneous policy decisions.

Beyond GDP

GDP is broadly considered a poor measure of welfare by many economists. In 2018, Joseph Stiglitz said the following:

“GDP is not a good measure of wellbeing. What we measure affects what we do, and if we measure the wrong thing, we will do the wrong thing. If we focus only on material wellbeing – on, say, the production of goods, rather than on health, education, and the environment – we become distorted in the same way that these measures are distorted; we become more materialistic”

One measure proposed by Jones and Klenow (2016) suggests taking into account a variety of additional factors – leisure time, inequality, and life expectancy. In their work, they start their analysis by noting the welfare issues of GDP with an example. In 2005, French GDP-per-capita was 67% of the US. By using just GDP, we could erroneously assume that life in France was a third worse. However, life expectancy in France was higher (80 vs 77 years in the US), hours worked were lower (Americans worked 877 hours per person, while in France, only 535 hours), and inequality was significantly lower. Clearly the GDP measure did not paint a full picture. Jones and Klenow decided to take the above factors into account when creating a single measure, which they refer to as ‘consumption-equivalent’ welfare. The idea is to convert all important welfare factors into a common unit of consumption-equivalent welfare (in the case of GDP, the common unit of valuing everything is currency). In order to create this unit of measure, the authors estimate values that enable them to compare, for example, how much extra consumption is 1 hour of leisure time worth. Naturally, these estimations will all depend on the country in question (1 hour of leisure in one country might be more valuable than in another). After creating this consumption equivalent welfare unit, they can compare countries directly. Based on their measure, they got the following results for 2007 (to interpret the table, Sweden has 91.2% of consumption-equivalent welfare of the US, which means Swedes have 8.8% less welfare than Americans).

This is just one potential approach to an all-encompassing single measure.

Hulten and Nakamura (2022) discuss other important features that need to be accounted for in welfare metrics, although they themselves do not specifically propose a measure. They argue that one key factor missing from all welfare measures is the ‘consumption technology’ level influencing our lives. They base their theory from Kelvin Lancaster’s insight that welfare depends on the characteristics of goods consumed and not on the goods themselves. This is best explained with an example: “a meal which is more than just the sum of the individual items of food consumed, but a complex interaction of various factors”. This complexity can grow even further as a dinner party will have even more welfare gains as it mixes a meal with a party. Therefore, Hulten and Nakamura argue that in order to measure welfare, we need to convert simple goods and services (measured by GDP) with some form of consumption technology (i.e. changing food into a meal). Examples of such goods and services that generate far more welfare than their GDP value are modern medicines due to their ease of access and use and the Internet due to its access to unlimited knowledge and even online dating, simplifying meeting people.

Conclusion

GDP is a useful measure, as long as its limitations are well understood. Using the measure within its specific context gives a lot of information about the state of the economy, but it does miss a lot of welfare implications. This fact has been picked up on by many government institutions, including the European Union, the OECD and the Bureau of Economic Analysis, that are now working on developing new measures, commonly referred to as ‘Beyond GDP’. Hopefully, these initiatives will come to fruition.

1. Recently in 2022, the US experienced negative GDP growth in the first and second quarters of 2022. However, a recession was not declared by the National Bureau of Economic Research (NBER), as they use other factors as well when declaring a recession. This decision caused significant media debates.

2. This was a conference of delegates from 44 allied countries to establish a regulatory system for the international monetary and financial order after World War II.

3. Note – this is not the same “investing” done by buying stocks on the secondary stock market. This activity does not get counted in GDP.

4. Nearly all central banks only have the goal of maintaining price stability.