Which is right: high-flying stocks or ho-hum GDP? It's one of this bull market's defining questions, and GDP recently lost some ground in this intellectual horserace, thanks to the very statisticians who calculate it. That's right, the BEA's bean counters recently determined they've missed a few over the years, due to the increasing difficulty of calculating inflation. For stocks, the discovery doesn't mean much-the findings are backward-looking, while markets discount the future. But it does provide more evidence that when markets and the economy (measured by government statistics) seem disconnected, markets are generally a more accurate-and timely-gauge.
Ever since Simon Kuznets spearheaded the effort to create national economic accounts in the 1930s[i]-basically creating GDP in the process-economists have been on a never-ending quest to perfect the measurement. From the start, Kuznets warned accounting for all productive activity is an impossible task-and the final measurement would inevitably be wrong.[ii] Determining what to count and where to classify it has only become trickier as technology improves and the digital economy mushrooms. As a recent book on the subject, Ryan Avent's The Wealth of Humans, points out:
"In the industrial era, a firm that bought a giant machine tool directly contributed to measured GDP. That counted as investment, which gets included in the national statistical accounts. In the digital era, investment in social capital[iii] does not register in the data. When a group of people comes up with a brilliant business model, that doesn't show up in GDP. When a firm reorganizes itself to take better advantage of digital technology, that doesn't show up in GDP." (pp. 131 - 132)
Couple that with GDP's other shortcomings, including imperfect inputs (e.g., not recognizing imports as domestic demand, counting government spending as always positive), imprecise measures (more below) and mathy scrubbing (like seasonal and inflation adjustments) that introduce their own set of biases-and it's clear the stat isn't the be-all, end-all arbiter of economic health that too many pundits make it out to be.
Most economists are well aware of all of this, hence the never-ending refinement and historical revisions. A few years ago, they recalculated GDP all the way back to the 1920s after deciding investment in research & development should count as business investment. Yet most coverage of GDP presumes it's an airtight, infallible measure. Slow GDP throughout this expansion fueled media skepticism, which in turn influenced broad investor sentiment. But perhaps this will change soon. Careful analysis by the agencies that compile GDP reveals they've overstated inflation for decades, thus underestimating "real" GDP, which is econospeak for inflation-adjusted.
Posing the question: "Are measures of economic growth biased?" economists affiliated with the Bureau of Labor Statistics (BLS) and the Bureau of Economic Analysis (BEA) concluded that real GDP growth may be understated by -0.4 percentage point annually from 2000 - 2015 due to better medical care and information technology improvement. Extant reporting, they argue, doesn't accurately estimate medical procedures' effectiveness and technology upgrades that provide consumers "more value for their money." So, for example, if a more expensive stent substantially reduces mortality from coronary heart disease, then a higher price shouldn't necessarily subtract from the value that adds to real GDP. It isn't actual inflation, like the price of cheese rising. It's a product improvement. Or, say, the recent widespread adoption of electronic stability control systems[iv] in cars also raises their prices, but is highly effective in saving lives and reducing crash severity. Is that inflation, or a better product that happens to cost more? Add it all up-tack 0.4 percentage point onto each year's growth rate and recalculate the compound growth of 1999's GDP through 2015-and the cumulative difference amounts to around $8.2 trillion of missing "value."[v]
We aren't saying it would be super bullish if all this-and then some for years post and prior-were added to GDP. Markets look forward. Prior revisions-like reclassifying R&D as an investment in 2012 and recalculating GDP's entire history-had no impact. As far as stocks are concerned, this stuff is trivia. But it does illustrate stocks' superior discounting mechanism. Sufficiently liquid markets price in all widely known information and opinions.
While GDP has lagged its average growth rate during this cycle, the S&P 500 has roughly matched its 21% long-term average annualized return during bull markets. Economic fundamentals are only one driver, but it seems fair to say stocks weren't concerned by the specter of slower growth. They saw through the math. Again from Avent's TWoH:
"Intangible capital is becoming more important over time. In the 1970s, big firms were tangible animals. A recent analysis considered how much it would cost to duplicate the average firm on the S&P 500: that is, how much you'd have to spend to obtain the machines, buildings, technology, workers and so on that represent the visible components of a company. In the 1970s, the value of those components added up to more than 80 per cent of the firm's valuation. The rest of the valuation constituted what was then defined as 'dark matter': the stuff you can't just go out and buy. Today, however, these proportions of value are reversed. More than 80 per cent of the value of top firms resides in these intangibles - stuff that simply can't easily be accounted for; the buildings and salaries and all the rest of it are only a small chunk of what makes a valuable firm valuable." (p. 107)
Just because something isn't counted in GDP statistics doesn't mean it hasn't happened or doesn't matter. Standard-of-living improvements driven by corporate innovation over the last decade are amazing, whether or not they're officially considered "economic growth." And regardless of the GDP numbers, those improvements provide new sources of earnings, which is what you own when you own stocks.
[i] National Income and Product Accounts-from which GDP is derived-went into production in 1934 to determine the Depression's extent and then was more fully developed in 1942 (GNP's official appearance) to assess the economy's capability to gear up for fighting WWII.
[ii] His primary quibble was with GDP's excluding stay-at-home moms. Most folks wouldn't say unpaid housework and child/elderly care, for example, are unproductive-indeed they're critical for the economy to function-yet the "value" of such work is entirely subjective ... and unrecognized in GDP statistics.
[iii] Described as: "contextually dependent know-how, which is valuable when shared by a critical mass of people. The social capital of successful firms is increasingly the most important component of their success: the shared understanding of how the firm does what it does is more valuable than the machines it uses or the patents it holds." (p. 119)
[iv] Essentially souped-up ABS and traction control systems.
[v] As in, the differences between each year's GDP and woulda-been GDP, added up.
If you would like to contact the editors responsible for this article, please click here.
*The content contained in this article represents only the opinions and viewpoints of the Fisher Investments editorial staff.