Eurasia Press & News
In China and Elsewhere, Forget the Numbers—Look at the Lights
Winston Smith, the protagonist of George Orwell’s 1984, spends his days as an employee at the Ministry of Truth adjusting documents to conform to the fluctuating political needs of Big Brother’s regime. “Who controls the past controls the future, who controls the present controls the past,” Orwell wrote, drawing a connection between a government’s ability to influence what the public perceives as the truth and that government’s political survival. Although he leaned to the political left, Orwell was deeply wary of the tyranny coalescing in the Soviet Union, which sought to bend reality to its needs. Joseph Stalin had a penchant for manipulating photographs, airbrushing out figures whom he had purged. He also wanted to rule untrammeled over the field of official statistics. Stalin had the bureaucrats in charge of the 1937 census arrested and executed after their findings displeased him; the initial results suggested that the Soviet population had not grown as much as expected in the preceding years, mostly as a result of the 1932–33 famine caused by Stalin’s collectivization of agriculture.
Governments of all stripes remain preoccupied by such statistics, and many leaders feel compelled to cook the books. In a December 2022 address, Chinese President Xi Jinping claimed that his country’s GDP was expected to exceed 120 trillion yuan ($17 trillion, around $6 trillion less than that of the United States) for the year. If true, this figure implied an annual growth rate for the economy of approximately 4.4 percent, much higher than the 3.3 percent growth rate that independent forecasters had expected. It also suggested that the Chinese economy is well on its way to matching its American counterpart, at least in absolute terms, and could overtake the U.S. economy in as little as a decade. Such a feat would be the culmination of what is already the greatest economic success story of recent times and would further increase China’s geopolitical heft.
But there is good reason to doubt the magnitude of Beijing’s economic achievements. The Chinese government and those of other autocracies are especially prone to inflating statistics regarding their performance. Of course, leaders in democracies try to burnish their records in all sorts of cynical and even deceitful ways, but their statements tend to face greater scrutiny and resistance. Autocrats can lie much more easily. My research, which parses satellite images of nighttime lights in order to provide a more accurate measure of economic activity, finds that autocracies habitually overstate their economic success. China’s GDP growth, in truth, is not as high as its leaders insist, and the country is not as close to catching up with the United States as is commonly presumed.
COOKING THE BOOKS
China’s statistics have long inspired doubt. Academics have debated for years whether the numbers produced by the Chinese government on topics as diverse as air pollution and workplace safety should be trusted. Beijing’s decision to abruptly end its strict “zero COVID” policies at the end of 2022 prompted questions about the credibility of its official COVID-19 mortality figures, which remained impossibly low.
When it comes to GDP figures, there are good reasons to be skeptical. Chinese local officials produce data on economic growth within their jurisdictions, which central authorities then weigh in decisions about whether to promote those same local officials. The inextricably politicized nature of these figures has forced many leaders to take them with a pinch of salt. The unreliability of some of China’s official economic statistics seems to be an open secret among high-ranking officials, as evidenced by former Premier Li Keqiang’s acknowledgment in 2007 that railway cargo volume, electricity consumption, and loan disbursements were more credible measures of economic activity than blunt GDP figures.
Unlike official GDP growth statistics, nighttime lights cannot be manipulated.
To be sure, this impulse to manipulate official records to paint a rosy picture is not exclusive to nondemocracies. In the early 2000s, several member states of the European Union engaged in creative accounting in order to artificially comply with supranational fiscal rules. At the end of the same decade, inflation measured through online prices in Argentina was three times larger than the official estimates, a discrepancy that was not observed for other large economies in Latin America. In Colombia, the head of the National Statistical Agency resigned in 2004 amid allegations that he had been pressured by the government of then President Alvaro Uribe not to release the results of a survey on security perceptions, a sensitive topic for a government centered on law and order.
There are, however, several important differences between the distortions that occur in democracies and those that occur in autocracies. Democracies tend to allow public officials to speak up against the government without fear of reprisals and, more generally, create an environment that facilitates scrutiny and accountability. Officials in democracies and autocracies may both have incentives to overstate government performance, but leaders in democracies have a harder time getting away with it. A well-functioning system of checks and balances allows for the examination of official statistics by political opponents, judicial institutions, the news media, and the public at large. Moreover, the protection of basic civil liberties facilitates the publicization of evidence of misreporting.
These institutional constraints are largely absent in authoritarian regimes, which allow for greater control and manipulation of information. As the economist Sergei Guriev and the political scientist Daniel Treisman argue in Spin Dictators, the desire and capacity to manipulate information is the defining feature of present-day autocrats. For instance, the Soviet government initially denied the catastrophic nuclear accident at Chernobyl in 1986 and admitted the magnitude of the disaster only when radioactivity measures skyrocketed in other countries. In recent times, Venezuelan President Nicolás Maduro and Turkish Prime Minister Recep Tayyip Erdogan have put a tight muzzle on all news media in their countries. Russian President Vladimir Putin’s government has engaged in a massive misinformation campaign regarding its invasion of Ukraine in 2022.
THE VIEW FROM ABOVE
In recently published research, I examined whether autocracies are more prone to overstating economic growth than democracies by analyzing both official GDP figures and the spread of nighttime lights as recorded by satellites. Scholars have begun to use such lights as a proxy for economic activity. An economic expansion leads to an increase in private and public infrastructure (more houses, more factories, more roads) and higher electricity consumption. All of these factors produce more lights in the night, which the U.S. Air Force’s defense meteorological satellite program can measure at a highly granular level in publicly available data sets. This form of data has several key advantages for the study of economic activity: it covers much of the globe, it allows for easy comparison across locations and over time, and it can be aggregated to different units of observation, such as villages, provinces, and entire countries. Most important, unlike official statistics regarding GDP growth, nighttime lights cannot be manipulated.
I parsed the available data on nighttime lights in 184 countries between 1992 and 2013, comparing changes in the relative brightness of a country with its reported GDP data. Autocracies reported higher GDP growth figures than democracies did for the same amount of growth in nighttime lights, overstating yearly GDP growth by a factor of 1.35 relative to democracies. (This means that when the true growth rate is 1.0 percent, the authoritarian government reports 1.35 percent, or when the true growth rate is 10 percent, it reports 13.5 percent.)
These exaggerations of GDP growth are not constant across time or space. I found, for instance, that the discrepancy between nighttime lights and GDP is not apparent in developing countries during years that they benefited from concessionary loans and grants from the World Bank’s International Development Association. But once their gross national incomes grow beyond the point where they are eligible for these programs, they begin to inflate their statistics. This makes sense. Governments that derive a relatively large share of their revenue from the fact that they are poor have a weaker incentive to overstate how well their economies are doing. But once they no longer have much to gain from being poor, they seek the consolation of inflated numbers.
Governments exaggerate their GDP growth to a larger degree when their economies are underperforming relative to the rest of the world. Last year was by all accounts a difficult year for the Chinese economy, following three years of strict pandemic-related restrictions. It may be a coincidence, but just as China’s political elite gathered in Beijing for the 20th Congress of the Chinese Communist Party, the country’s National Bureau of Statistics announced a highly unusual delay in the publication of the official GDP figures. In his December address, Xi reported the robust figure of 4.4 percent GDP growth; if that figure is reduced by the factor of 1.35, as suggested by my research, China’s real growth rate would stand at 3.3 percent, the rate that was predicted by independent analysts.
REAL NUMBERS
Based on official figures, autocracies tend to grow at higher rates than do democracies. Between 1992 and 2013, countries classified as “not free” by Freedom House grew at an average aggregate rate of 85 percent, whereas countries classified as “free” grew at an average aggregate rate of just 61 percent. Nine out of the 14 countries with the highest GDP growth during this period were “not free.” But these long-run figures change once they are adjusted based on the estimates derived from my study of nighttime lights: “not free” countries grew at an aggregate rate of 55 percent during the period in question, while “free” countries grew at a rate of 56 percent. These figures suggest that the autocratic tendency to overstate GDP growth—not actual economic performance—accounts for most of the difference in official growth rates between democracies and autocracies in recent decades.
These findings also shed light on a question that occupies many analysts: when will China’s GDP overtake that of the United States and make China the largest economy in the world? Recent estimates have suggested that China could reach that milestone as soon as 2035. Despite the country’s unprecedented economic growth over the past 40 years, my research suggests that many current estimates may be excessively optimistic, since they do not account for autocrats’ habitual overstatement of GDP growth. Although China’s economic miracle is undeniable and the country’s economy seems on track to one day overtake that of the United States, its leaders will likely try to speed the arrival to this milestone by relying on bloated numbers and hopeful accounting.
