Fudged Numbers Won't Make the U.S. Grow Faster

The Trump administration has some very optimistic economic targets. It should resist messing with the statistics.

Hey, hands off the statistics.

Source: Myloupe/uig/getty images

The Soviet Union had a reputation for presenting fake economic statistics to the world, in order to avoid having its dysfunctional system look embarrassing in front of its rivals.

But as the decline and collapse of the Soviet Union illustrates, it’s impossible to conceal poor economic performance forever. There are too many alternative data sources that will eventually show red flags. If output is inflated, measures of electricity use and rail traffic will probably catch it. If inflation is understated, people can just go to the store and take their own surveys of prices. Economic sleuths are numerous and competent, and eventually the truth will out.

The danger is that in the short term, slanted data can cement support for bad policies or a failing regime. In the time it takes for the economic detectives to find the truth, five years of stagnation could turn into 15, or a modest housing bubble could become a gargantuan one.

The problem of falsified data is especially acute in China. Local officials get rewarded based on economic growth, so they have an incentive to exaggerate their regions’ performance. A whole lot of small lies will add up to big lies at the national level. The people at the top are well aware of the problem, but may themselves be shy about reporting bad numbers that could spark capital flight or asset price crashes.

The West has usually had reliable, high-quality economic statistics. This reinforced its reputation as the free world -- accurate information on government performance complemented democracy, because it allows the people to throw out any leaders whose performance is unsatisfactory. Though the power of U.S. presidents to harm or help the economy is often exaggerated, this system worked well.

The reliability of U.S. data was proven by repeated challenges. For example, after the Federal Reserve ramped up monetary easing efforts during the Great Recession, many thought inflation would increase. When government statistics showed inflation still low, some questioned the accuracy of those numbers. But the economic sleuths were on the case. Researchers at the Massachusetts Institute of Technology launched the Billion Prices Project, which aggregated prices from online retailers. The BPP found that though countries such as Argentina routinely underreported their inflation rates, the U.S. government’s data was remarkably accurate.

This record of unbiased objectivity should be preserved. The Donald Trump administration will be very tempted to start messing with government numbers to suit its own ends. Trump has made some pretty extravagant economic promises, including 4 percent growth and a big reduction in the trade deficit. The president has limited power to actually make those things happen, so there will naturally be a temptation to jigger the numbers to make it look as if the goals are being met. In addition, the administration’s tendency to fight with the bureaucracy and to embrace alternative facts bodes ill for its commitment to the independence of government data-collecting agencies. Trump himself has also accused government unemployment numbers of being phony -- asserting that the real jobless rate is as high as 42 percent (the most cited measure shows it’s 4.8 percent).

Already there are some worrying signs. The Trump administration has pushed the Council of Economic Advisers to up its forecasts to account for the pro-growth effects of Trump’s policies. Forecasts are not the same as data, of course -- they rely heavily on theory, not just numbers, so they’re always dependent on assumptions. And every administration probably leans on the CEA to some extent. But the new administration seems to have even less compunction than normal about pushing bureaucrats to spit out rosy forecasts.

More worrying, the administration is also reportedly considering changing the way trade deficits are calculated. The revision would count goods that are imported into the U.S. and then immediately resold outside of the country -- so-called re-exports -- in the import numbers, but not in the export numbers. That’s a bad idea -- because imports are netted out of gross domestic product to avoid double-counting, this change would end up undercounting economic output. It would also overestimate the trade deficit -- something that enters the country and then immediately leaves again shouldn’t add or subtract to the balance of trade.

Finally, Trump’s hiring freeze will make it harder for government data-collection agencies to do their necessary work. Accurate data depends on an impartial, independent bureaucracy.

The administration should think twice about this sort of thing. If it pushes too hard to fudge official economic numbers, it may gain a temporary political advantage, but in the long term it’s going to get caught. It would be a shame if the U.S. developed a reputation for falsifying numbers in the same way that the Soviet Union and China did. Reliable data is part of the lifeblood of a healthy democracy. If we start fudging our numbers, we could eventually feel it in our pocketbooks.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

(Corrects eighth paragraph to indicate the Council of Economic Advisers rather than the Congressional Budget Office.)

    To contact the author of this story:
    Noah Smith at

    To contact the editor responsible for this story:
    James Greiff at

    Before it's here, it's on the Bloomberg Terminal.