President Trump’s firing of the Bureau of Labor Statistics chief Erika McEntarfer attracted much criticism, as government statistical agencies are claimed to be impartial. Such claims of impartiality should always be regarded with deep suspicion, as they generally conceal a noxious political agenda. When you look at the BLS’s statistical methods, its agenda becomes apparent. Ever since the Boskin Commission report of 1996, the BLS has reported consumer prices using “hedonic pricing,” a highly opaque methodology that biases reported inflation sharply downwards. That falsification inflates reported growth and living standards, thereby giving us a falsely optimistic view of economic reality. Ms. McEntarfer may be blameless individually, but the BLS and its sister agency the Bureau of Economic Analysis have for decades needed to be taken apart and rebuilt using sound economic principles.

Professor Michael Boskin’s (1945- ) Commission in 1995-96 was set up to find a way to massage consumer price data downwards to give the Fed and other policymakers an excuse to run with lower interest rates and report higher “real” growth numbers. It found spurious reasons to report the CPI as being overstated by 1.3% per annum and proposed a method, “hedonic pricing” by which that “bias” could be removed. Its rationales for the CPI being overstated were:

1. that consumers substitute less expensive goods when relative prices change (which does not alter the fact that prices have changed)
2. that consumers change outlets (which should be handled by reflecting those outlet changes in the prices sampled – in any case a switch to Internet buying has increased prices rather than reducing them)
3. quality improvements are not measured – but nor are the much more common quality deteriorations, as sellers switch their offerings to cheap Chinese rubbish that falls apart at first use
4. New Product bias – new products are not included in the market basket immediately

To remove the alleged biases in CPI statistics – which before the change, contrary to Boskin’s spurious claims, represented pretty closely ordinary people’s cost of living – the BLS invented “hedonic pricing” by which quality improvements would be subtracted from the CPI. This suffered from two problems: it could not be measured accurately and it did not take account of quality deteriorations such as Maytag washing machines no longer lasting 50 years without repair. The problems were made much worse by the BLS’s solution in the tech area: it counted quantitative improvements in chips’ transistor numbers (by a law first propounded by Gordon Moore (1929-2023), the number of transistors in a typical chip used in a tech product doubles roughly every two years or did until recently) and used them as a proxy for qualitative improvements in the tech product’s functionality.

The BLS hedonic pricing model adjusts for increased chip size each year and applies that factor as a reduction in prices in the tech sector, then re-scales the tech sector so that it represents a constant (or gradually increasing) percentage of the total goods basket. In 1996-2025, the number of transistors on a chip increased by a factor of roughly 20,000; that factor, re-scaled every year, has been applied to reduce reported CPI increases. In reality, Microsoft Word or Excel bought today, most people’s principal day-to-day application of chip technology other than social media, may possibly be twice as good as in 1996; they are certainly not 20,000 times as good.

Thus, CPI growth has been “hedonically” suppressed by a gigantic factor that represents nothing. Whatever overstatement there may have been in the CPI before 1996, there is now an understatement of at least 1% per annum. Since price information is used to calculate “real” increases in GDP and “real” increases in wages and living standards, and “real” increases in productivity, those figures have been correspondingly inflated in 1996-2025.

This has many effects on the economic realities we have all been taught by the media and the government. The productivity growth “acceleration” of 1997-2004 was an artifact of hedonic pricing; in reality productivity growth remained roughly constant during that period at the level of the early 1990s, well below that of 1948-73. Since 2004, productivity growth first slowed to a crawl and then since 2011 has been negative, as more and more input has been required for each unit of real output. There are three reasons for this stagnation: (i) regulatory bloat, which has greatly worsened since the George W. Bush administration, (ii) the influx of ever larger armies of unskilled and incompetent foreigners into the workforce and (iii) insane monetary policy, which since 2011 has artificially given us negative real interest rates and caused massive over-investment in unproductive office buildings, tech rubbish and a gigantic stock market bubble.

Income growth figures are also affected. The corporatist think tank blob has recently circulated a graph showing median real wages taking off around 1996 after a flat period in the 1980s. This makes no economic sense and is utterly wrong. In reality, median real wages have stagnated since the 1970s, as the bubble in asset prices has siphoned off more and more wealth to the top 0.01%. Median real wages are now lower than in 1996 and men’s median real wages are below 1973 levels (when many women did not work). GDP growth has been very sluggish since 1996, even more so since 2011, and much reported GDP growth has taken the form of asset price increases and malinvestment, which will reverse once the current asset bubble bursts. In short, the BLS’s hedonic pricing scam has given us an entirely false picture of the last 29 years’ economy, resulting in the encouragement of insane Keynesian leftist policies by both the government and the Fed. Reality needs to dawn!

I wrote in March about the other principal problem with government statistics, this one perpetrated by the Bureau of Economic Analysis, and dating back to the intellectually dodgy New Deal economist Simon Kuznets (1901-85). The BEA propagates and we use Gross Domestic Product as a proxy for output when we should be using Gross Private Product, in which government output is valued at its proper level of zero. In addition, the BEA’s figures for government output are artificially deflated and private output inflated by pretending Medicare and Medicaid, the fastest growing programs, are part of the private sector.

The BLS itself admits its unemployment statistics, the principal focus of adverse attention by President Trump, suffer from increasing difficulty with the declining willingness of cellphone users to participate in surveys – this also makes the BLS’s revisions after the first estimate larger in many cases than the first estimate of job increases. In addition, the survey uses two different methodologies, the establishment survey and the household survey, whose results have increasingly diverged.

Those problems may not be the BLS’s fault; however, the artificially low reported unemployment rate is a deliberate choice, made to make the figures look better and reduce public pressure on the government of the day. The true position is indicated by the declining labor force participation rate, which peaked around 1998 and has since entered an inexorable decline – and no, women are not increasingly choosing to be housewives, which kept the participation rate low in the 1950s and 1960s. The labor force participation rate will only recover when it is made worthwhile for domestic workers to choose employment over a life of disability benefits, opiates and idleness, a problem that President Trump is addressing.

One especially encouraging recent statistic is that foreign-born employment declined 1.01 million between January 2025 and July, whereas native-born employment increased by 2.46 million. This is a huge sign of the success of President Trump’s repatriation policies, and of the robust health of the U.S. economy (government employment also declined during the period, another excellent sign of economic health). In the previous year, January 2024 to 2025, incidentally, foreign-born employment increased 1.93 million whereas native-born employment increased only 0.77 million on a base five times as great as foreign employment. That should have been a giant warning bell to President Biden, had he been capable of comprehending it.

There is a huge job to be done at the BLS and the BEA, in making their statistics fit for purpose, and President Trump is to be commended for attempting that Herculean task. Once that is done, I am confident that, over the squawks of the legacy media, they will report a strong economic surge under President Trump, with living standards increasing and the labor participation rate rising as unskilled jobs are paid at a level that is attractive to “discouraged” men and women, making work more attractive than Fentanyl. The new tariffs will help here also; they will make it too expensive to import low-quality rubbish from China and more attractive to recruit and train a decently paid workforce to produce quality goods and supply top-notch services in the United States.

As a first priority, could the new BLS head please get rid of hedonic pricing. It’s a deceptive scam!

-0-

(The Bear’s Lair is a weekly column that is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that the proportion of “sell” recommendations put out by Wall Street houses remains far below that of “buy” recommendations. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)