Redefining GDP

Gross domestic product (GDP) is a standardised and widely used metric to assess the economic health of a nation. The appeal of the statistic, according to the 2009 report by the Commission on the Measurement of Economic Performance and Social Progress (also called the Stiglitz-Sen-Fitoussi Commission), is that it is “a single headline figure allowing simple comparisons of socio-economic performance over time or across countries”.

Despite its popularity, GDP’s shortcomings have been the subject of controversy and intense debate since its inception almost eight decades ago. Much of the debate has centred on what is included in the measurement, such as reconstruction following a natural disaster or environmental catastrophe, and what is not: for example, domestic housework or elderly relative care (unless, of course, it is done by hired help).

As a 2004 article in the Organisation for Economic Co-operation and Development (OECD) Observer summarises: “[GDP] measures income, but not equality, it measures growth, but not destruction, and it ignores values like social cohesion and the environment.”

However, many economists, including Shanta Devarajan, acting chief economist at the World Bank, argue that the main issue stems from extrapolating societal trends from an indicator meant solely to measure the value of final goods and services produced in an economy.

Growth or welfare?

The headline criticisms surrounding GDP as the sole measure of economic success have focused on its failure to account for the overall wellbeing of society and income distribution, particularly in the wake of the global financial crisis. Thomas Piketty’s 2013 book, Capital in the 21st Century, which looks at wealth and income inequality in Europe and the US since the 18th century, added fuel to the fire.

“Piketty’s book, among other discussions, has prompted a greater focus on inequality,” says Mr Devarajan. “Many people are looking at the US economy’s reasonable growth, for example, and yet see increased inequality, which throws up many questions. There is also the environmental dimension, such as climate change, and an increased public policy focus on these issues. The public looks at GDP and finds that it doesn’t capture what they are concerned about.”

Goolam Ballim, chief economist at Standard Bank, agrees that the debate about GDP has increased at a time when the depth of inequality has exposed material fault lines in key developed markets. “The prevalence of poverty and inequality, spanning both emerging and developed markets, has infused the global economy with renewed political and geopolitical risks,” he says, pointing to Donald Trump in the US, Brexit in the UK and anti-immigration impulses across Europe that have stoked nationalist sentiment. “This is raising questions within the historical methods of [GDP] measurement.”

Yet Mr Ballim argues that GDP remains a reasonably credible measure for less developed economies and emerging markets. “In these markets, the nominal level of income and income growth can be a good steer of a household’s purchasing power, especially for basic goods and services,” he says. “However, in advanced economies, where essential needs are largely satisfied and a greater proportion of income is allocated to discretionary items, there is room for additional measures that explain welfare and wellbeing.

“Pragmatism dictates that it is not necessarily that GDP as a metric has reached sunset, but that it needs to be complemented by a variety of other metrics. A resulting portfolio approach could then inform policy in a more constructive way.”

Complementary metrics

Many of the large international institutions and agencies are developing supplementary indices to address GDP’s shortfalls, without abandoning the metric altogether. For example, in 2017 the World Economic Forum launched its Inclusive Development Index, which “identifies 15 areas of structural economic policy and institutional strength that have the potential to contribute simultaneously to higher growth and wider social participation in the process and benefits of such growth”.

The OECD has its Better Life Index, which contends that “there is more to life than the cold numbers of GDP and economic statistics”; whereas the United Nations has its Human Development Indicator, with data dating back to 1990.

The World Bank’s Changing Wealth of Nations (CWN) report tracks the wealth of 141 countries between 1995 and 2014. CWN is unique, according to Mr Devarajan, because it calculates the stock of a country’s wealth – including produced, human and natural capital, and net foreign assets – rather than the flow of wealth, as GDP does. He believes CWN is a complementary indicator that helps to address some of the criticisms aimed at GDP.

But with a proliferation of alternatives, achieving consensus on which one to move to is a major challenge, according to Diane Coyle, Bennett professor of public policy at Cambridge University, and author of GDP: A Brief But Affectionate History. “GDP is like a technical standard – everyone uses it and understands it in the same way, and it makes it easy to compare countries, relatively speaking,” she says. “There won’t be real change until there is enough agreement about what the new standards should be.” 

Joe Brusuelas, chief economist at RSM US, an audit, tax and consulting services firm, agrees with Ms Coyle that, to date, there is no consensus around what should augment GDP. “There is no one method that currently is favoured, or gaining traction, in such a manner that there is a possible point of convergence to improve our conceptual understanding of economic growth, vitality and welfare,” he says.

The digital shift

In addition to the debate around GDP’s inadequacy in reflecting general welfare and sustainability, another issue is attracting interest among economists: digitalisation. Mr Brusuelas believes this global trend is more likely to cause rapid change in the near term than the debate over inequality, raising the possibility of a radical rethink in how GDP is measured.

According to Janet Henry, global chief economist, at HSBC: “Technology is allowing the provision of various goods and services in different ways to the past – for example, streaming Spotify rather than buying CDs – and, in some cases, the provision of goods and services that before technological transformation were not possible, accessible or were exceptionally expensive to provide. Much of the latter involves using smartphones, which are now affordable in large parts of even the least developed countries.”

For example, online educational courses can now be delivered all over the world and some are provided for free. “As such, these do not show up in the expenditure measure of GDP, but over time will hopefully improve productivity,” says Ms Henry.

It is not just the ability to deliver a service for free or at a much lower price, but also the fact that the quality of services has improved in the digital economy. Again, this is not reflected in GDP estimates, says Mr Devarajan. “While many aren’t entirely free because advertising is incorporated, services are still probably underpriced. But what is the value of the improved connectivity that Facebook provides for free?” he asks.

Productivity puzzle

Mark Zandi, chief economist at Moody’s, reports that the rating agency has done some work on estimating the shortfall in GDP growth related to the shift to a digital economy. “Slow GDP growth, particularly in this economic expansion, is related in part to accounting for the pace of technological change and the resulting improvement in an array of consumer and business technologies,” he says, speaking to Mr Devarajan’s point.

For example, Moody’s research estimates that measurement errors have shaved roughly 0.2 percentage points off annual GDP growth during the current US economic expansion. “In any given year, that is no big deal, but over a nine-year period it is almost two percentage points of GDP, which is real money and makes a big difference,” says Mr Zandi. “While it still doesn’t explain the total shortfall of productivity growth in this expansion period, it goes part way to explain what is going on and why things don’t seem to be adding up.”

Likewise, Mark Cliffe, chief economist and head of global research at ING, argues that many digital services, free goods and services, and intangibles are not being captured, while others are badly measured. ING’s research, soon to be published, indicates that GDP is understating growth and overstating inflation. “At some point, there must be a broad-based re-examination of how we measure progress,” he says.

RSM’s Mr Brusuelas believes the inability to capture the essence of the digital economy is the single greatest challenge in constructing estimates of national economic activity and vitality. “The inability to accurately measure the free apps, for example, alters estimates of growth and inflation in much the same way that it does not fully capture the benefits of increasing quality of goods over time,” he says. “If one assigns a zero value to goods that are free and perfect, one is not accurately capturing the added value from such goods, nor does one accurately capture national welfare of the standard of living in an economy.” He also believes that both growth and productivity in the US economy are happening at a much faster pace than the official statistics would indicate.

This failure to truly measure the value-add of digitisation is contributing to the “productivity puzzle”, says Ms Coyle. “For example, instead of buying servers that go out of date and are costly to maintain, many businesses are using a cloud computing service provider and getting access to the latest technological advances, such as artificial intelligence, while paying less and getting better services. Yet they are being told their productivity is flat,” she says.

Mr Brusuelas adds: “The real economy firms see gains in productivity every day. They see the depths and the order of magnitude of change in terms of what they can produce based on the integration of advanced technology into the production of goods and provision of services.”

Revision on the cards?

Ms Coyle reports a increase in research to improve the kind of statistics collected on the economy. For example, she is part of a research centre funded by the UK Office for National Statistics called the Economics Statistics Centre of Excellence. “We have a work programme looking at how to improve the fit between GDP and the structure of the modern economy. Much of this is quite challenging and complicated, such as determining the quality of different services and accounting for it in a price index,” she says.

Both Mr Zandi and Mr Brusuelas point to recent efforts by the US Department of Commerce’s Bureau of Economic Analysis (BEA) to adjust for the rate of improvement in different products and services. The BEA’s benchmark revision indicated that the economy between 2012 and 2017 had increased 0.4% faster than had been previously acknowledged, which makes a substantial difference in a $19,000bn economy.

“One of the fascinating things about the report was that by going back and looking at cloud computing and mobile phones, the BEA found out that total business investment increased by a whole percentage point on average for each year from 2002 to 2017,” says Mr Brusuelas. “One does not have to be an economist or a policy expert to imagine how this might have impacted the central bank policy path had this data been available in real time.”

This is important, Mr Zandi says, because of the heated debate around how fast or slow economies have been growing, which ties into the political debate and resulting policy choices. “For example, if GDP growth is underestimated, so is productivity growth,” he says. “The productivity growth shortfall, particularly in the longer run, is critical to our understanding of how well the economy is performing or not, and what policymakers should be doing about it.”

Mr Brusuelas agrees. “Just by focusing on a select number of areas, we are learning much more about how fast the economy is growing, how fast it is capable of growing, where productivity is at, which then informs a whole host of policy judgements that are made primarily by fiscal and monetary authorities. The reason why this is important is because what we can measure tends to get managed.”

New technologies

While the economics community is still at an early stage in the application of new approaches, ING’s Mr Cliffe believes that the digitalisation that is distorting current GDP metrics also potentially provides a way of gathering and analysing data. “The emergence of big data, artificial intelligence and machine learning is an emerging field in macroeconomics as well. It opens some interesting possibilities, such as real-time measures of activity,” he says. He points to ‘nowcasting’, which is the tracking of current economic performance using high-frequency, even real-time, data.

Ms Coyle agrees that there is a lot of scope to use emerging technologies for data collection, for example using satellite data or mobile phone data to get a much better handle on growth in low-income countries than from official GDP statistics. “Many countries don’t have the capacity or resources to collect GDP statistics like in the US or UK. But new techniques like web-scraping, text analysis and machine learning will eventually help with collecting and processing data,” she says.

Most expect that utilising big data will transform the ability to appraise economic health and activity. Mr Devarajan reports that the World Bank is looking at whether big data can be used as an alternative to surveys to collect economic information. “The jury is still out, and we need a few more years of research, but I think it is promising,” he says.

But much information of that kind is held privately, for example by the large technology platforms. Ms Coyle supports a discussion around data as a public good. “Official statistics of the conventional sort are usually required to be provided by law and businesses must provide the information. But there isn’t a legal or standards framework in the domain of digital data. I think a discussion around a digital common that creates public statistics is another part of the agenda,” she says.

Mr Zandi agrees, while Mr Brusuelas argues that to change the way GDP is measured will take the utilisation of big data and the requirement of firms to move towards an open-source environment, in which almost all data can be captured and disseminated to policy-makers in a near real-time environment.

However, Standard Bank’s Mr Ballim believes that GDP will remain an anchor metric for many years to come: “While the shortcomings inherent in using GDP [as a measurement of economic strength] have certainly been exposed, it is a case of refinement and complementary metrics, rather than necessarily the absolute isolation of GDP as an indicator.”