Elevator pitch
Gross domestic product (GDP) is the key indicator of the health of an economy and can be easily compared across countries. But it has limitations. GDP tells what is going on today, but does not inform about sustainability of growth. It does not measure happiness, so residents can be dissatisfied even when GDP is rising. GDP does not consider environmental factors or reflect what individuals do outside paid employment. It might increase in times of military conflicts and after natural disasters or terrorist acts, as the loss of property is not counted. Hence, complementary measures may help to show a more comprehensive picture of an economy.
Key findings
Pros
GDP provides a headline number to focus on the economic condition of a country.
GDP is the mainstay of modern macroeconomic analysis and essential to policy creation.
Because GDP conforms to international measures of production, it is an objective indicator that is not subject to national political pressures.
In most cases, GDP uses well-established international accounting principles.
GDP concepts are modified as the economy changes.
Cons
GDP does not capture welfare or human well-being.
GDP may not be a strong basis to predict economic growth in times of high uncertainity.
As international accounting standards are slow to change and require international consensus, GDP is slow to reflect changes in the world.
GDP excludes non-market activity, notably human capital creation in that sector.
Important indicators such as ecosystems accounts, measures of happiness, sustainability, or inequality are not included in GDP, other than through satellite accounts.
Author's main message
GDP is the single most important indicator of the overall state of an economy and should be retained. It is a standardized measure across countries, and is relied upon by policymakers to determine whether action is needed to reduce inflation, or to stabilize or grow the economy. However, GDP should be one of a suite of measures. It is unlikely that accounts and measures supplementary to core GDP accounts will give a good sense of a country’s future, in part because national income accountants try to avoid making risky, assumption-laden, or normative choices.
Motivation
Gross domestic product (GDP) measures a country’s economic activity, or total output. As such, it is a “neutral” measurement, in that it makes no judgments about what is good or bad economic activity—it is simply the market value of whatever is currently produced in a country. Thus, according to GDP, even natural disasters, military conflict, and acts of terrorism can increase growth. For example, following the May 2008 earthquake that struck Sichuan Province in China, in which at least 80,000 people died and which left the area in ruin, Chinese authorities maintained that, “on net,” this led to economic growth [1]. Property destruction resulting from natural disasters is not deducted from GDP, although reduced expenditures, and personal and business income are subtracted. Weapon manufacture is also counted as part of GDP. In addition, the military may attract some individuals as it has better employment prospects than some of the alternatives, thereby raising GDP by increasing income and buying power.
So, given the above, why is GDP considered to be the key indicator of the health of the economy? GDP does well in capturing what it intends to measure. It focuses mainly on the more easily measured and tangible elements of the economy: what is bought (consumption and investment) and spent (government). But as a comprehensive measurement of a country’s overall state of economic and environmental health and social well-being, it leaves something to be desired.
Discussion of pros and cons
The relevance of GDP to policy
For most individuals, one headline number, typically quarterly or annual growth in GDP, is more than enough to focus on when considering the economic condition of a country. Is the economy getting better or worse, and is growth higher or lower than usual? Analysts, researchers, and policymakers, however, need to understand and appreciate the details and trends behind the GDP figures to a much greater extent. For example, growth often fluctuates, and it is the longer trends of a year or more that are most important to capture.
The US as an example of GDP policy process
The policy making process in the US, for example, typifies, in general terms, what happens in the policy making arena for most countries in the world. The Chair of the US Federal Reserve, who is responsible for monetary policy, carefully examines with her advisors GDP components and other measures, such as labor force statistics, to inform her decisions. The Council of Economic Advisors (CEA) and the National Economic Council (NEC) are bodies that advise the US president on economic policy. The Joint Committee on Economic Policy (JCEP) and the Joint Committee on Taxation (JCT) are two of many US Congressional Committees that closely monitor growth in GDP, as they consider legislation that can have an impact on the economic condition of the country. The US national economic accounts, first published in 1937, were created under the leadership of Simon Kuznets following the Great Depression, when it was realized that there was scant data on which to base economic policy [2]. In fact, it has been suggested that the creation of the US GDP accounts has moderated swings in economic activity because these accounts led to a better understanding of current conditions [2].
National income accountants, who estimate GDP, follow a rigorous and detailed set of accounting standards in compiling GDP: the System of National Accounts (SNA) [3]. The SNA principles are followed by national statistical offices (NSOs) around the world. Although the US has another set of standards embodied in the National Income and Product Accounts (NIPAs), the NIPAs closely conform to the SNA standards in order to ensure comparability, although its format is quite different. Developing countries may not have the resources, in terms of qualified experts and finance, to fully implement the SNA, as its manual is over 600 pages long and very complicated, even for the expert.
In most countries, the national income accountants are not allowed to engage in policy creation, discussion, or analysis. If national income accountants engaged in such activities, they might be tempted to, or be perceived to, fashion GDP to serve their personal or political preferences. In the US, the agency that estimates GDP—the Bureau of Economic Analysis (BEA)—has no political appointees. Political office-holders, such as the US president or secretary of commerce, who may have access to the GDP numbers through the CEA before they are released, are not allowed to comment on the numbers in any way until after they are released to the public. Strict security measures are enforced to ensure there are no early leaks of the GDP number or of its components. BEA employees have to achieve the highest standards of impartiality and secrecy in estimating GDP. Several countries around the world, albeit with some variation, reflect this structure and process.
How measuring GDP has changed
Elements of the economy that are tracked in GDP have changed significantly since the 1930s. Countries determine what GDP changes to make and when to make them. For example, for the US, there have been many changes in definition, classification, source data, and methodological approaches. Since the mid-1980s, many analysts have been concerned with price or quality changes, such as hedonic price indexes for computers (that capture the changing value of component parts and characteristics) and the implementation of indexes to better capture inflation and quality-adjusted changes. Other changes include the classification of software and research and development as investment, separation of government expenditures into consumption versus investment, and measurement of implicit services provided by property and casualty insurance and by commercial banks. Each year in the US, at the end of July, an annual revision of GDP occurs; some definition, classification, source data, and methodological changes occur at this time, but major changes occur every five years as part of a comprehensive revision. Data are revised back through time to ensure comparability of historical figures. Over time, because of an initial different set of accounting standards, the US system has become more similar to the SNA. The EU publishes its own set of methodology guidelines, which also may not be identical to the SNA because of timing or other issues.
The SNA changes much less frequently because it requires the agreement of a large number of countries through an advisory process. An important concern is the extent to which poor or less-developed countries can implement proposed changes. The SNA was created under the leadership of Sir Richard Stone and was first published in 1947. Since then, six new versions have been published, in 1953, 1960, 1964, 1968, 1993, and 2008. The 2008 version is an update of the 1993 SNA manual [4].
Alternatives to GDP as a comprehensive measurement
There have been significant criticisms of GDP over the last years. The Stiglitz–Sen–Fitoussi Commission, which popularized the buzzword “beyond GDP,” is clearly the primary example of critical evaluation, was created by the then President Sarkozy of France in 2008 [5]. Its report calls for substantial innovation in economic accounts, including GDP. A strong emphasis of the report is a switch from production-based measures to measures focusing instead on income and consumption. The report concludes that material living standards are more closely associated with measures of real income and consumption than with GDP. Sometimes production expands or contracts without a commensurate expansion or contraction in material living standards, as GDP does not include income flows in and out of a country or adjust for depreciation. In addition, there may be consequential differences in consumers’ prices versus the price of GDP. It was thought that such a switch, with underlying household detail and with companion wealth and income distribution measures, would facilitate the construction of measures of income distribution, inequality, and poverty.
The Stiglitz–Sen–Fitoussi Commission therefore recommended that material living standards should be tracked on a household basis, with accounts considering ingoing and outgoing financial payments, including payments and any “in-kind” benefits or services. Non-market activities were thought to be best captured through “satellite accounts” to the GDP accounts, with services provided, time used, and home-production goods included, particularly as the latter can be very important for developing countries. In contrast, the report recommends indicators as opposed to monetary or numerical measures, in almost all regards, when considering societal well-being and sustainability (i.e. economic, environmental, and social sustainability, and their interlinkages). The exception is a call for surveys that ask individuals to subjectively and objectively describe their quality of life and priorities and measures that capture social capital, inequities, and the relationships between personal quality of life dimensions. However, statistical offices were asked to provide enough information so that an index could be constructed to capture these well-being features. The conclusion is that sustainability should be presented in a “dashboard” of indicators, in which monetary measures should be given a secondary role because of the difficulty of constructing them. The environment and environmental considerations are also suggested to be part of this dashboard.
In addition to “beyond GDP,” there are a number of alternative measures of GDP that have been developed as a result of a perceived overreliance on it, or as a preference for yardsticks that emphasize economic growth to a lesser extent, or even not at all. The UN’s Human Development Index (HDI), which has been published since 1990, covers 200 countries. It currently includes gross national income (GNI) per capita, although formerly it included GDP per capita. GNI is an important companion to GDP, from the “other” side of the accounts, as it captures actual or imputed income to individuals, businesses, and other entities. GNI is an SNA concept (as is GDP), which excludes net taxes on production and imports. Unlike GDP, GNI includes income flows in and out of countries. These flows are the main difference between GDP and GNI. Accordingly, a country’s GNI is typically larger than its GDP if income inflows are greater than income outflows. The HDI also includes education and health as subcomponents, specifically life expectancy at birth, expected years of schooling, and mean years of schooling [6]. The recent HDI report focuses on the impact of work on human development, as well as presenting inequality-adjusted HDI measures. It defines work broadly to include salaried jobs, unpaid, voluntary, and creative work.
Happiness is another possible measure. The most recent World Happiness Report, published in 2016 by the Sustainable Development Solutions Network, describes happiness measures for 156 countries. Six factors, comprising levels of GDP, life expectancy, generosity, social support, freedom, and corruption, are sub-components of the happiness measure [7]. Generosity is measured by: recent donations; social support (having someone to count on in times of trouble); and freedom (perceived freedom to make life decisions). Survey respondents are asked to give their subjective ratings of happiness on a scale from zero to ten; it has been shown that these rankings can be linked to variations in the six factors listed above. The latest report is an update in a series that started in 2012. As well as describing the overall country happiness measures, the recent report describes the levels of happiness inequality among countries and within countries and regions.
Nevertheless, however important, neither the HDI, the World Happiness measures, nor the SNA satellite measures receive anything like the attention from policymakers that is paid to GDP. This is because the predominant focus for policymakers is on economic growth, as opposed to other, equally, if not more, important (but certainly complementary) considerations.
Revise GDP?
Some researchers and commentators concentrate on possible revisions to GDP as opposed to alternative measures. In a recent book, the history of the development of GDP is summarized, the difficulty of coming up with an accurate measure is described, and some revision suggestions are made, from the standpoint, nonetheless, of a belief in the GDP measure as a “bright light shining through the mist” [8].
The overall conclusion of this book is that a new way for thinking about “the economy” has to be designed, for three reasons: (i) its increasing complexity; (ii) the increasing share of components that are difficult to measure, specifically services and intangibles, and particularly for advanced economies; and (iii) the importance of quantifying depreciation and depletion of assets to assess sustainability [8].
The International Association for Research on Income and Wealth (IARIW) conference in 2015 (“W(h)ither the SNA”) brought together many practitioners in the art of constructing GDP and other researchers. There were at least three points of view expressed. One view was that GDP should be substantially replaced and changed, as it does not measure welfare in a way that includes environmental and social measures (such as income distribution, inequality, and the contribution of unpaid work). Another view called for keeping the basic framework “as is,” but to add to it similar measures as listed above, plus to add to it constructs reflecting globalization and the greater importance of intangibles (such as research and development, brand equity, and human competency). The other view was to leave GDP “as is,” without substantial changes or additions. Also, in part, the discussion at the conference highlighted the frustration of some with the difficulty and slow pace of making substantial changes to the SNA; but on the other hand it demonstrated the extent to which developing countries have consistently found it difficult to implement.
Benefits of using GDP “dashboards”
Dashboards can present a variety of indicator measures (often significantly more than five), and sometimes these measures are weighted to construct a summary measure. Perhaps the best example is the HDI, which covers 200 countries [6]. It has three major components with equal (geometric mean) weights: GDP, health, and education. However, even its summary measure is frequently criticized by those who say its weights are arbitrary. Decisions about weights, by nature, are always subjective, rather than objective.
The compendium produced by the OECD’s “Better Life Initiative” allows users to create their own weights for 11 topics of well-being to form an overall indicator based on their own subjective preferences [9]. It concentrates on the well-being of individuals in up to 38 countries, rather than the macroeconomic conditions of their countries. There are 24 indicators underlying the 11 topics, in the areas of housing, income, jobs, community, education, environment, civic engagement, health, life satisfaction, safety, and work-life balance.
Dashboards are difficult to use to evaluate progress. However, unlike GDP, which implies that growth is always good and the supreme target of a country, other critical aspects of a country are at least being recognized.
GDP, HDI, and World Happiness measures
To what extent do the HDI and World Happiness measures differ from GDP? The Illustration shows the percentage change in the HDI versus the percentage change in GDP, in constant prices, between 2013 and 2014, for 173 countries. Percentage changes are shown in the graph because most attention is paid to GDP rates of growth—rather than to the absolute level of GDP or to GDP per capita—by policymakers and by the public press. Visually it shows, as statistical tests confirm, that there is little relationship between the rates of growth of the two different measures. However, there is a strong correlation between the overall HDI and its education component [6].
Figure 1 charts the 2016 World Happiness measure, an average over 2013–2015, versus nominal GDP per capita, for 139 countries. In this figure, Qatar is not included as its happiness measure at 141 is so much higher than the happiness measure of any other country. Although there is greater dispersion in the points in Figure 1 compared to the Illustration, a logarithmic equation fits the Figure 1 data reasonably well, with the World Happiness measure explained by nominal GDP per capita.
Other alternatives to GDP
The construction of other types of accounts, besides HDI and happiness accounts, could also be useful. For example, human capital accounts based on household production or non-market activities could be an offshoot of micro household-level data on income and consumption. Human capital accounts, if human capital is constructed with a lifetime income measure which recognizes the contribution of individuals throughout their whole working lifetime, are an important component of sustainability considerations too.
Considerations of “time poverty,” particularly for women who historically have borne the bulk of the responsibility for caring for the young, the infirmed, and the elderly are also important societal considerations. Caring time is an implicit obligation today, and will be in the future, as parents and carers age. Interest in well-being, which is usually more broadly defined than happiness, has also substantially increased in recent years, in part because of the Stiglitz–Sen–Fitoussi Commission recommendations.
Limitations and gaps
An advantage of GDP is the extent to which it is based on tangible measures, such as the quantity of goods produced. Other measures, such as happiness, depend to a greater extent on components that are more difficult to conceptualize and assess. However, this dichotomy is becoming less clear, as GDP itself depends more and more on the quantity of services produced and intangible elements that are estimated using more subjective methodologies than the quantity of goods produced. Additionally, even the “quantity” part of the quantity of goods produced is a less certain measure, as it is quality-adjusted to a significant degree, particularly for technology-intensive products, such as semi-conductors. Continuing and intensive research by both government and private researchers is therefore needed to maintain GDP’s accuracy as a measure of economic activity in a rapidly changing world.
Summary and policy advice
In spite of the existence of some alternative measures, there is comparatively little press coverage or attention by policymakers on any measure other than GDP and labor market measures, such as the unemployment rate. One of the press coverage exceptions is the Gross National Happiness Measure of Bhutan. Its four pillars include good governance, sustainable socio-economic development, preservation and promotion of culture, and environmental conservation [10]. Also, in recent years, much attention has been paid to the construction of well-being measures, with happiness being one component of well-being.
An important question to ask is whether the fact that GDP attracts so much attention means that it is the “right” measure? The answer to that is “no.” However, it does highlight the difficulty in replacing it with another measure, or a suite of measures. A second question is whether GDP should be replaced? Again, the answer is “no.” GDP should not be replaced; it is no more than what it purports to be, i.e. a measure of the value of production. It would be best if there could be a more widespread understanding of its limitations, though this would be difficult to accomplish for any statistical measure. Also, more frequent updating or revision of the SNA would be ideal.
A third question is whether GDP should be complemented by a suite of other measures? And, if so, how could these other measures gain sufficient attention and traction to become credible and accepted? One answer would be to release an alternative or complementary measure along with the GDP release, at least once a year. The issuing GDP agency could be assigned the responsibility for the construction of the other measures, in order to give them credibility, or they might be calculated by another body, even an international one. If the GDP agency does not construct the alternative measure, it is possible that a list of a few alternative measures could be included in the body of the GDP release, without the agency necessarily implicitly indicating their approval of the alternative measures or how they are constructed.
National income accountants may not be the ideal specialists to comment on alternative measures. In some countries, particularly in those for which there is one national statistical agency—as opposed to the triumvirate that exists in the US (i.e. the Bureau of the Census, the Bureau of Economic Analysis, and the Bureau of Labor Statistics)—other measures are already constructed by the same agency, but GDP still garners the lion’s share of attention, with the possible exception of labor market measures, as previously noted.
Simultaneous release of GDP with its alternative complementary measure(s) might increase the attention paid to the other measure(s). An international task force could be formed (e.g. by the UN Statistical Commission) to consider how to best increase the amount of attention paid to alternative measures. This would require the commitment of dependable funding from international organizations—which would of course be problematic during times of budgetary stress.
Should all countries focus on one alternative measure, such as a set of household-level income and consumption accounts? The attention given to Thomas Piketty’s recent analysis of income distribution [11] suggests that this could be a credible case, particularly in view of the fact that income distribution, inequality, and poverty numbers do seem to attract attention.
The World Bank produces a range of poverty measures for over 200 countries [12]. International agencies certainly need to be responsible for the construction of such a set of household-level income and consumption accounts, as many developing countries often lack sufficient expertise, data, or funding to keep up even with the current version of the SNA. A new manual would be needed, much simpler and shorter than the SNA, that is regularly updated to keep up with the changing nature of the world.
In summary, there are three principal policy recommendations. First, to keep GDP as it is, as a measure of production, but to: (i) update SNA methodology more frequently; (ii) allow for a more simplified SNA version by countries with limited resources; and (iii) add a brief text to all GDP releases that emphasizes what is and is not included in GDP and what it represents, e.g. it is not a measure of well-being or welfare. Second, to construct an income and production measure that: (i) has a household base, allows for considerations of individual well-being, poverty, income inequality, as well as perhaps “time use” and “time poverty”; and (ii) is made available along with a GDP release in the release text or a simultaneous release at least once a year. And finally, to convene an international group that includes national income accountants to consider how to raise interest in and knowledge of alternative measures to GDP.
Acknowledgments
The author thanks two anonymous referees and the IZA World of Labor editors for many helpful suggestions on earlier drafts. The analysis and conclusions expressed in this article are those of the author and not necessarily those of the China Center for Human Capital and Labor Market Research of the Central University for Finance and Economics in Beijing China.
Competing interests
The IZA World of Labor project is committed to the IZA Guiding Principles of Research Integrity. The author declares to have observed these principles.
© Barbara M. Fraumeni
The System of National Accounts (SNA)
The Stiglitz–Sen–Fitoussi Commission
Source: Stiglitz, J. E., A. Sen, and J.-P. Fitoussi. Report by the Commission on the Measurement of Economic Performance and Social Progress (2009). Online at: http://www.stiglitz-sen-fitoussi.fr/documents/rapport_anglais.pdf
