In Praise of GDP
Gross Domestic Product remains the best measure of overall welfare and efficiency available to economists
The quest to measure and comprehend welfare has long posed a formidable challenge to economists. Over the past century, numerous approaches have been proposed to gauge welfare, each providing a unique lens. Some approaches emphasize production, others income inequality, and a growing number pay heed to environmental outcomes. Amid these varied proposals, one measure has been a consistent anchor: gross domestic product, or GDP.
Despite its popularity among economists, investors and journalists, this long-standing yardstick of economic prosperity is also frequently criticized and even maligned. But are the criticisms fair? Is GDP really an unreliable representation of welfare?
The purpose of this article is to explore these questions further. The following sections will delve into the history of GDP, unpack the criticisms, and contemplate its strengths. We will examine attempts to refine and enhance GDP, for example by integrating considerations such as environmental sustainability into the measure. In the end, however, the conclusions is that GDP does indeed offer significant value. Production is the source of national wealth and prosperity, and GDP per capita is one of the better indicators of cross-national welfare, given that it is so strongly correlated with many indisputable aspects of human well-being.
Moreover, many of the criticisms of GDP turn out to be overstated. In some cases, what seem like detriments even turn out to be benefits. For example, the measure unambiguously excludes certain aspects of well-being, but it may be that economists’ attempts to compare differential aspects of well-being—such as by “monetizing” certain nonmarket phenomena—obfuscate more than they illuminate. Hence, GDP is better off without incorporating these kinds of revisions.
That being said, GDP does have shortcomings that should not be overlooked. GDP is best thought of as a short-run measure of economic health, and its focus on an aggregate statistic can mislead when important developments occur at disaggregated levels of analysis. Nevertheless, GDP’s persistent relevance should not be attributed to the dubious honor of being the least bad of all the relevant alternatives. The measure offers fundamental insights not just into the magnitude of a nation’s wealth but into its citizens’ welfare as well.
The History of GDP
GDP is a measure of the economic output of a nation. It is defined as the total value of all final goods and services produced within a country’s borders during a specified time period (typically a year). “Final” goods and services are goods and services purchased for their final end use—that is, those that will not be resold or processed further. Final goods can be contrasted with intermediate inputs into products: for example, a fuel injection pump that might be used as a component in an automobile. The value of intermediate goods is excluded from GDP to avoid double counting.
A nation’s income and production are essentially two sides of the same coin, and they are measured in three ways: adding up income streams, adding up expenditures or looking at the value added at each stage of production. These three approaches to national income accounting are correspondingly known as GDI (gross domestic income), GDP, and GVA (gross value added), and in theory they should all add up to the same amount.
In practice, however, because of measurement issues, there can at times be significant divergences between the various measures. For example, in late 2022 and early 2023 there was some debate about whether the U.S. was in a recession. According to the GDI measure, there were multiple quarters of negative growth: the informal definition of a recession. According to GDP, on the other hand, growth was positive during the same period.
GDP is commonly reported in several forms, including “real GDP,” which is adjusted for inflation so that comparisons can be made across time, “nominal GDP,” which is unadjusted for inflation, and “GDP per capita,” which expresses national production on a population-adjusted basis.
The concept of GDP dates back to the 1930s and was first formulated by the economist Simon Kuznets. As part of an effort for the U.S. National Bureau of Economic Research, Kuznets devised a system of accounts that would measure the nation’s economic performance during the Great Depression. This led the U.S. Commerce Department, in consultation with Kuznets, to create the National Income and Product Accounts, which still exist today.
Kuznets’s original measure was a precursor to gross national product, or GNP, which counted the production of goods and services by Americans whether they were operating domestically or internationally. The switch to GDP, with its focus on domestic production (regardless of whether that production is from domestically-owned companies), occurred in 1991.
In a report to the U.S. Congress in 1934, Kuznets famously warned against using his new production measure as a catchall benchmark of economic progress. “The welfare of a nation can, therefore, scarcely be inferred from a measurement of national income as defined above,” he noted, referring to his net national product measure. By the end of World War II, however, GNP was well on its way to becoming the standard tool for measuring a country’s economic growth. National income statistics received a major boost at the Bretton Woods Conference in 1944, an international gathering aimed at designing a new global monetary system and promoting international cooperation and economic growth. World leaders wanted a method to track the economic progress of countries for the purposes of reconstruction efforts, and GNP was well-suited for that purpose.
Kuznets went on to win a Nobel Memorial Prize in Economic Sciences for his contributions to the statistical methods used to measure a nation’s economic growth.
Criticisms of GDP
Despite the ubiquity of GDP as an indicator of a nation’s economic health, it is almost as well known for its shortcomings, particularly its failure to account for many aspects of human, environmental and animal welfare. GDP measures economic activity as positive, regardless of its social or environmental impact. So, for example, market production that results in significant external damages to the environment is still counted as making positive contributions to national output. The same goes for economic activity that decreases animal welfare: GDP does not count welfare costs to caged chickens on factory farms or to rabbits or rats used in laboratory experiments. Instead, animals are regarded as commodities.
Similarly, expenditures made in response to natural disasters or to prevent crime contribute to GDP but do not necessarily signal that society is any richer. Some criminal activity gets included in GDP: for example, money criminals spend on legitimate market goods in the service of black market enterprises or other illegal activities. Corporations that defraud customers or investors may also contribute to GDP. And government spending contributes to GDP irrespective of whether that spending accomplishes anything.
GDP excludes consumer surplus—the benefit to a consumer over and above the price they pay for a product—and excludes certain intangibles in the digital economy, such as the value of apps or other technologies that are available for free. And GDP does not consider income inequality. Two countries might have similar levels of GDP but vastly different distributions of income and wealth. (Simon Kuznets proposed the hypothesis behind the “Kuznets curve,” which posits that inequality might grow for a time as a nation becomes richer before eventually declining.)
GDP also overlooks some forms of nonmarket production, in the form of household work or volunteering, as well as leisure time. So, for example, if you cook a hamburger for yourself at home, it is not included in GDP, but if you purchase the hamburger from McDonald’s, it is. Cleaning your apartment yourself is not included in GDP, but hiring a maid to perform the exact same chores is included. A vacation spent at the beach isn’t reflected in GDP despite providing value to the vacationer.
Finally, GDP fails to account for quality improvements over time. Thus, by virtue of having a lower sales price, a single television set sold today might contribute less to GDP than one sold 50 years ago, even though the newer product is vastly superior.
GDP as a Measure of Efficiency
The sheer number and variety of criticisms leveled at GDP demonstrate how a measure of production and social welfare often deviate. Though GDP certainly has limits as a welfare measure, there are nevertheless some compelling reasons to keep using it. First and most obviously, a measure of production may be useful in its own right. Even as a welfare measure, however, GDP has its advantages.
Consider, for example, the surplus value a consumer attains above and beyond the price paid for a product. Because consumer surplus is not reflected in the price of the market transaction, it does not get directly tabulated in GDP. Yet it is debatable whether this omission is of significant concern at the aggregate level of production. To illustrate, if a consumer is willing to pay $5 for milk but need only spend $3 because competition has driven the price down, the resulting $2 surplus can be allocated elsewhere. When the consumer spends it, it is incorporated into GDP.
This indirect method of counting consumer surplus highlights how problems found at disaggregated levels of analysis can sometimes be rectified at a more aggregate level. In this sense, adding consumer surplus value to GDP often results in a kind of double-counting problem. However, GDP does not account for all surplus that manifests itself in nonmonetary forms. For instance, if industrial activities threaten a species of turtle, this state of affairs may impose a cost on an environmentalist who would be willing to pay to avert this negative outcome. Yet if there is no market for saving the turtle, this “cost” does not diminish the environmentalist’s purchasing power, and thus there is no indirect impact on GDP. For this reason, there is an important distinction to be made between externalities that affect products priced in markets and those that come solely in nonpecuniary form.
One compelling reason to track GDP is that fostering new markets via entrepreneurship can effectively address concerns about externalities, public goods and other market failures that critics of GDP frequently highlight. Consider what would happen in the turtle example if a new market were created for turtle protection. Say I am willing to pay for turtle preservation: if the cost of protecting turtles is less than the minimum price I am willing to pay, an astute entrepreneur should be capable of delivering the service of turtle protection and charging me accordingly. Once the new market is established, costs and benefits associated with turtle preservation will be accounted for in GDP. This market for turtle preservation may of course produce countervailing market failures elsewhere. Nevertheless, one market failure has been addressed.
A related issue concerns the production of public goods such as ideas, which are frequently identified as one of the main catalysts of economic growth. Ideas, like any public good, can be underprovided by the market, constituting another form of potential market failure. In that sense, there is nothing particularly special about ideas. Like other unpriced elements of the economy, ideas do not contribute to GDP, but this can change when they give rise to marketable products. Some ideas, such as the idea for a funny joke, may never be commercialized at all and therefore never influence market production. As in the turtle example, this represents a kind of market failure—a failure to seize an opportunity, since someone might have been willing to pay for the joke at below cost. However, if a shrewd entrepreneur can monetize the joke—for instance, by selling the service of stand-up comedy—the market failure is addressed, and the joke idea will be reflected in GDP.
Again, the creation of a new market can, via spillover effects, reduce the scope of market and nonmarket activity elsewhere. Because an ideally efficient market would encompass markets for every conceivable outcome affecting human welfare, GDP will reflect a measure of overall efficiency in the aggregate. In a world of complete markets, all benefits and harms imposed on humans by any activity whatsoever would be accounted for in GDP. For this reason, GDP can serve as a measure of the completeness of markets and the benchmark by which society assesses the extent to which market failures have been rectified.
Quality and Time Adjustments
Perhaps the most serious complications arise for GDP as a result of the quality improvements (or degradations) that GDP struggles to account for over time, difficulties that are exacerbated by the limitations of inflation indices. Comparing this year’s GDP to last year’s can provide a reasonable—although admittedly still imperfect—picture of economic growth. But it is not at all clear whether such comparisons are meaningful across longer time spans.
Imagine, for example, an uncomplicated economy that generates four oranges, three pizzas and five television sets in a year. In the following year, the same economy produces five oranges, four pizzas, six television sets and a piece of furniture. It is clear that the second year’s production exceeds the first, but the degree of improvement remains somewhat ambiguous. Inflation indices are used to convert GDP measures into a common basket of the same real goods and services. However, the piece of furniture in this example introduces a completely new element, making an exact measurement of the extent of real economic growth undetermined.
With each passing year, direct comparisons become increasingly difficult. What if we rewind 30 years and find that the economy I described earlier was producing two wooden toy rocking horses, a steak dinner and a basket of bananas? How would this economy compare to today’s, whose production set is entirely different? Relatedly, can we even consider a phone from 30 years ago to be the same product as one made in the internet age with touch-screen technology?
It should be evident that a statement like “GDP grew by 3 percent last year” is pretty reasonable, assuming measurements are precise. However, a statement like “GDP grew by 250 percent since 1975” is far less reliable—in fact, it is potentially meaningless. The man on the street’s intuition about economic growth since 1975 may be just as reliable as the economist’s measurements.
It is clear, then, that GDP primarily serves as a barometer of short-term economic health. Nevertheless, the fleeting nature of the GDP measure’s relevance does not diminish its utility for providing snapshot comparisons across short time spans or between different geographic locations at unique moments in time.
GDP as a Measure of Well-Being
Although GDP per capita will never fully capture everything along the spectrum that constitutes human welfare, it nevertheless exhibits a positive correlation with many factors that are directly connected to welfare. As GDP per capita increases, many well-being indicators typically improve, providing a useful means of comparing the general welfare of different countries.
One critical factor with which GDP per capita shows a positive correlation is health outcomes. It has been consistently observed that a higher GDP per capita is linked to superior health care systems and healthier populations overall. A practical illustration of this relationship can be seen in data published by the United Nations, the World Health Organization and other international bodies. According to such reports, countries that have a higher GDP per capita tend to have longer life expectancy. In richer countries, health care infrastructure is typically better funded and more accessible, leading to a healthier population overall. Additionally, richer nations tend to have lower rates of infant mortality, another key marker for health care quality and accessibility.
A further area in which GDP per capita demonstrates a strong correlation with well-being is educational attainment. It has generally been found that countries with a higher GDP per capita have better-funded education systems. This financial backing often translates into a higher literacy rate and higher levels of educational attainment among the population. Data from UNESCO supports this, as does the Barro-Lee Educational Attainment Dataset. Nations with greater economic output per person tend to have better educational outcomes for their citizens.
GDP per capita is also closely associated with political stability and with quality of governance. Data from the World Bank’s Worldwide Governance Indicators provides evidence of a positive correlation between GDP per capita and factors such as the rule of law and control of corruption. In essence, nations with a higher GDP per capita are more likely to have effective, transparent governance and a stable political climate. This stability can, in turn, foster more economic growth, creating a positive feedback loop.
To some extent, each of the previous correlations could be explained by reverse causality: for instance, perhaps higher life expectancy, better health care or greater educational attainment raise GDP rather than being raised by GDP. The correlations could also be affected by omitted variables: perhaps external factors raise both GDP and these welfare measures. Nevertheless, it is striking how strongly correlated GDP is with measures of welfare. One study by Stanford University economists Charles Jones and Peter Klenow found that, across countries, GDP per person has a correlation of 0.98 with an index the authors constructed that includes variables thought to be associated with well-being. This suggests that GDP per capita is an almost perfect predictor of welfare as gauged by these authors’ chosen variables.
Finally, GDP per capita provides a robust measure of material well-being. Wealthier nations and individuals have the means to purchase a wider array of goods and services, and this directly enhances living standards. In a country with a high GDP per capita, citizens are more likely to have access to high-quality housing, diverse food options, advanced technology and a variety of leisure activities. These material goods and services improve quality of life and contribute to overall welfare.
Attempts to Improve GDP
Given the limitations of GDP, it should not surprise us that alternative metrics have been proposed over the years. Early efforts, including the Genuine Progress Indicator and the United Nations’ Human Development Index, sought to encompass broader aspects of societal progress by integrating factors such as environmental impacts and health indicators. The latter index was introduced by Pakistani economist Mahbub ul Haq and was further developed with the help of Indian economist and Nobel laureate Amartya Sen.
In more recent times, focus has veered toward “green accounting” constructs such as the System of Environmental Economic Accounting from the United Nations. This approach aims to factor environmental costs into measures of economic progress, weaving together traditional production indicators with ecological data and measures of “natural capital,” which is an environmental counterpart to physical capital in the marketplace.
Such measures are sometimes colloquially referred to as “Green GDP” measures, a banner designation for environmentally-adjusted measures of GDP. The U.S. government has experimented with versions of the National Income and Product Accounts that include natural capital and has plans to incorporate such accounting into broader economic statistics by 2036. Other initiatives, such as the World Bank’s “Adjusted Net Savings,” aim to correct for educational attainment and environmental costs.
Although broad adoption of these statistics is a long way off, one exception, in the context of climate change, is the Dynamic Integrated Climate-Economy model, conceived by Nobel laureate William Nordhaus. Nordhaus has long been an advocate of green accounting, and his DICE model of optimal economic growth incorporates environmental impacts from the effects of greenhouse gas emissions. The model, as well as several similar models, has been used in the context of U.S. federal regulatory policy.
Juxtaposing natural and physical capital, as is done in some of these measures, is a task fraught with complexity, however. Market-based physical and human capital generate financial returns that can be reinvested and spur further economic growth. Natural capital, for the most part, does not (although exceptions exist). The complications of integrating the different forms of capital mirror many of the challenges facing economists who perform cost-benefit analysis, which entails attempts to “monetize” the value of nonmarket amenities, allowing them to be compared with market activity that involves the exchange of money.
Nonpecuniary benefits, however, are not equivalent to monetary increases in income, due to the unique characteristics of money. This should be evident from the earlier discussion involving consumer surplus. As a result, the temporal trajectory of the value of nonmonetary benefits is often radically different from that of benefit streams produced from capital stocks producing financial returns, and these differences would need to be accounted for before integrating natural capital in national income statistics.
Some nonmarket benefits are completely fleeting, such as witnessing a shooting star or a beautiful sunrise. These benefits operate more like consumption in the market economy. Other benefits might be ongoing as people enjoy the “outputs” of nature year after year in perpetuity. Natural capital in this sense looks a lot like investment. However, these benefits stemming from natural capital often cannot be reinvested—because they come in the form of an experience and not money. Such benefits will usually not generate compounding growth unless they have spillover effects on markets elsewhere (such as through improved worker productivity or health). A final form of natural capital relates to elements of nature that are unpriced themselves, but that nevertheless contribute to economic production, such as fisheries or timber.
The shift toward green accounting frameworks faces significant challenges with respect to comparing the benefit streams from these different forms of natural capital. The lack of wider adoption of green GDP alternatives should not be surprising given the overall inability of economists to adequately address these challenges.
Not All Production Is Created Equal
A limitation of GDP only briefly alluded to thus far is its focus on aggregate production, which can conceal facts pertinent to welfare at a more disaggregated level of analysis. For example, the specific composition of production in terms of public vs. private spending, investment goods vs. consumption goods, or in terms of goods production vs. services production, also has important implications for welfare.
Consider, for example, an individual who chooses to spend income on items of questionable value, such as tobacco products, junk food or lottery tickets. Should a measure of this person’s income be adjusted downward to reflect poor spending choices? On the one hand, the person’s income is not actually lower because of what it is spent on. Further, spending decisions that are arguably wasteful will tend to have repercussions for future income, influencing measurements at a later date.
At the national level, an overabundance of GDP dedicated to corruption, profligate government spending or excessive consumer spending could be a harbinger of looming problems. Do we sit idly by, waiting for the inevitable crash to show up in GDP? Or should we instead proactively seek economic indicators that will provide early warning signs?
There are arguments to be made on both sides of this issue, but one solution is to give greater prominence to disaggregated measures of spending and production. The libertarian economist Murray Rothbard recommended measuring “gross private product,” which represents what remains out of production after government has taken its share. The implication here is that private spending is somehow more productive—in the sense that it is voluntary—while government spending is imposed on people in an involuntary manner involving coercion.
Focusing on aggregate GDP also misses the fact that saving and investment are likely to yield greater long-run benefits than consumption, yet investment and consumption dollars are weighted equally in GDP. Indeed, a nation that prioritizes saving may appear to have lower living standards for a time because of its temporarily lower consumption levels. This misses the fact that forgoing consumption in order to invest yields economic growth and enables more consumption in the future.
Looking at shares of GDP also allows for meaningful comparisons across long stretches of time. However, there may be a double-counting problem to account for with investment dollars. Consumption measured in future periods of GDP reflects investments made in past ones. Some of the value of the future consumption might have been priced into the value of the initial investments, which suggests that downward revisions of conventional measures may be in order. On the other hand, if one is only considering national income at a singular point in time, this concern might be of little consequence.
Also important is the distinction between goods production and services provision. Goods-producing industries sometimes, though not always, correspond with more capital equipment. Manufacturing involves substantial investment in machinery, technology and reliance on infrastructure. Conversely, many service-oriented jobs, such as in retail or hospitality, employ labor to facilitate consumption. Intangible services, such as lawyers’ or accountants’ services, also are likely to be consumed compared to production of tangible goods, which have to be transported with trucks and equipment and stored in physical buildings. These items are more likely to act as stores of value.
Industries that require physical capital will have a longer-run multiplier effect on the economy than those that are purely service based, though of course many types of service industries, like hotels and restaurants, also rely on physical capital. Thus, the important distinction is not between goods and services per se, but the extent to which production is consumed vs. invested, and the extent to which industries depend on physical capital as inputs or outputs of the production process.
What these examples highlight is not necessarily that GDP measures need to change. Rather, they point to the fact that economists, journalists and pundits should supplement GDP with other measures to provide a more comprehensive picture of the economic health of a nation. One alternative, gross output, is a measure of production that includes the value of intermedia goods. Gross output could potentially be a better measure of total nominal spending in an economy, which may be more suitable for the purposes of business cycle analysis. Ultimately, however, what matters for human welfare is the number of final goods and services, regardless of the number of exchanges involved in their production. Thus, GDP—the measure of final output net of intermediate outputs—is likely a superior measure to gross output for welfare analysis purposes.
Nevertheless, such supplementary measures are likely a better way to account for elements of nature than amending national income statistics through adjustments to GDP or efforts to monetize natural capital stocks and flows unpriced in markets. Other disaggregated, supplementary measures could include the degree to which spending is directed by a nation’s government, the degree to which a nation saves, quantified but nonmonetized measures of natural capital, and the degree to which production comes in the form of goods with concrete, tangible value and a presence in the real world. The production of tangible goods, and the benefits of physical infrastructure and capital in general, are likely to become increasingly important as the economy further bifurcates into the physical and virtual realms.
The Future of GDP
Despite numerous critiques and suggested modifications, GDP remains a robust and valuable tool for understanding the economic wellness and welfare of a nation. Many objections to GDP ironically underscore its merits, such as its emphasis on generating wealth through market dealings, on prioritizing physical over natural capital and on taking into account the value of ideas only if they are productive enough to be successfully commercialized.
A focus on GDP, with its stress on the production of goods and services sold in the market, also aligns with a growth-oriented outlook. What the world may need more than anything else at this juncture is physical production in the form of mining for essential minerals (such as those used in batteries); energy generation from natural gas, hydropower, wind, sunlight and nuclear power; and goods production generally to raise the welfare of the global poor. All these activities directly or indirectly show up in GDP.
While GDP may not perfectly capture all aspects of welfare, such as environmental sustainability or social inequality, it provides a reliable measure of the material prosperity that underpins many aspects of societal progress. GDP has its limitations, to be sure, and there is a need for disaggregated and alternative statistics to supplement it, in addition to continual improvement and refinement of the traditional metrics. But GDP continues to be the most effective tool economists have for measuring national welfare.
The importance of a healthy population, a clean environment and safe workplaces should not be understated. However, raw inputs such as these need to be combined in useful ways to benefit humanity. Markets are the mechanism that arranges such combinations and guides them toward serving people’s needs. Ideas are not enough, natural capital is not enough, even people are not enough to generate a robust, flourishing society. Instead, society needs to do something with the resources with which it is endowed to create tangible value that raises welfare. The best metric we have to measure how much we are doing remains GDP.
Once you get how our income-based labor force really works (that high profits depend on low wages), you will finally see and understand all the reasons why a global system that can match people to jobs, resources to communities, and everyday needs and demands to local production, consumption, and recycling operations is more sustainable and ethical than monetary methods practiced today, mainly because scientific-socialism, compared to scientific-capitalism, is actually more democratic; it values and views this very basic, very intuitive belief “universal protections for all” as both a human and environmental right.