The attacks in Paris on November 14 could have a significant effect on the world economy going forward. Geopolitical factors played a role in the Federal Reserve’s decision not to raise short-term interest rates in September. Similarly, the fear of further terrorism, resultant retaliation and the possible negative impact on economic activity may hold the Fed back in December. There is also the concern that both businesses and consumers will reduce their spending until they become more confident about future world stability. Finally, with ISIS striking targets beyond the Middle East, there is the clear option that a major coordinated military effort might be organized to subdue its forces in the Middle East. This could involve troops on the ground from a number of nations, including the United States, although to date President Obama has ruled that out.
The attacks by al-Qaeda on the World Trade Center in New York and other sites in 2001 were not followed by a series of incidents although there was widespread fear of further strikes at the time; the resultant escalation of security precautions proved effective. The U.S. equity market declined sharply in reaction to those assaults, but a month later, with world economies recovering from a recession, the indexes had recovered their lost ground. The market rallied on the Monday after the Paris attacks, even though the economies of the major developed countries have been showing weakness recently. Will the Paris attacks and the response to them slow business activity further?
Prior to the Paris carnage there was some good news in the U.S. We finally got a strong jobs number for October with a 271,000 increase in non-farm payrolls. Construction was the standout among sectors adding workers. The Federal Reserve was probably already planning to raise short-term interest rates in December and this report should encourage them. The Republicans are very likely working on a new message to replace “Obama is not a job creator; he is a job destroyer.” Average hourly earnings increased 2.5% on a year-over-year basis and many observers commented that wages were “accelerating.” Fed Vice Chair Stanley Fischer has argued that if productivity is not growing, “there is not a lot of force in the economy for real wages to rise.” So perhaps wages won’t “accelerate.” Before we get too carried away with the idea that the long period of wage stagnation is over and consumers will have more spending money available to keep the economy growing in 2016, I think it would be useful to look at the complex factors at work in in this economic cycle.
Goldman Sachs has identified three factors influencing their “wage tracker.” The first is labor market slack, which has depressed growth .6%, according to their estimates. The second is weak productivity growth, which has diminished wage growth by an estimated .3%. The third is low inflation, which has dampened wage growth by an additional .2%. Adjusting for these factors added together would produce wage growth of 3.2%, 1.1% above the current level, using their data. Recent business cycles have shown wage growth of 3.5% to 4%, so what is holding wages back?
The slow growth of the economy is clearly one factor. The Goldman estimate that the economy is growing about one percentage point below potential seems right to me and accounts for about half of the shortfall in wage gains. What is more interesting is the low level of productivity improvement. In the 1950s it was not uncommon to have productivity growth in excess of 4%. In the 1960s 3% was a frequent reading, but after the period 2000 to 2005 where productivity growth was 3% to 4%, recent data has shown 1%. The recovery cycles of previous periods have been characterized by forces that increased productivity. According to Strategas Research, “the me decade” and credit cards spurred growth in the 1980s and the computer in the 1990s. The problem with the current period is that housing, whose excesses were the principal cause of the recession in 2008-9 and which is playing an important role in the current recovery, does not improve productivity. It only improves the quality of life.
Looking ahead, I examined opposing views of productivity. One bearish view is espoused by Frank Veneroso, an economist whose work I have followed for twenty years. Veneroso starts with the statement that productivity growth in this cycle has been abysmal, averaging only .5% over the last five years. He sees no reason in history or theory for it to change in what he views as a mature economic expansion. In eight of the ten cycles since 1950 there has been a surge in labor productivity in the early part of an expansion and decay in the latter part. One of the exceptions was the 1990s, when capital equipment investment in technology caused a late cycle increase in productivity. This is not happening now. Many observers explain the current low productivity data by saying the economy is facing “headwinds.” These include: (a) a lack of business confidence in future demand; (b) a decline in entrepreneurship except in certain technology-related areas; and © a lack of willingness by financial institutions to provide loans to start-ups. These conditions are unlikely to change soon.
According to Veneroso, valuations in the technology sector may be reaching an extreme point, particularly for promising companies that are not yet public. Businesses have been using leverage to increase earnings per share and borrowing has been expanding at almost twice Gross Domestic Product growth, thereby putting business debt at an all-time high. Mergers and acquisitions, share buybacks and leveraged loans are approaching the 2007 peak. This data suggests that the headwinds are not impeding entrepreneurial activity and innovation so there is no reason to expect a significant gain in productivity.
Goldman Sachs has interviewed two Northwestern professors on the productivity issue. Robert Gordon explains the reason for low productivity is that “we’re using software and computers that are very similar to the ones we used ten years ago.” Gordon may be yearning for the true sea changes in productivity growth like those caused by the steam engine, electricity, the automobile, air travel and air conditioning. The computer is clearly in that category but follow-on innovations have been less dramatic. He comments that some people think social networking is as important as indoor plumbing. He also cites the slow roll-out of digital access to medical records as a case where there is potential to increase productivity but we haven’t taken advantage of it. I would add MOOCs (massive online open courses), which offer the ability to educate people remotely but have not gained the traction I would have expected by this time.
Joel Mokyr is on the other side: “product innovation has been particularly pronounced during the last 20 years and if that is the case, productivity statistics systematically under-measure the rate of technological progress.” Mokyr makes a distinction between product innovation and process innovation. If you can produce the same goods with less labor and less capital, productivity improves. That’s process innovation. Product innovation is harder to measure. That can improve your life without showing up in the productivity statistics. Antibiotics and anesthesia are examples. So is an automobile that will last longer and require fewer repairs. He sees a significant opportunity in 3D printing. Unlike some who believe low productivity is caused by “headwinds,” he believes the wind is behind us. A third economist, Alan Blinder of Princeton, wonders whether the prevalence of free services on the Internet inflates output growth without concomitant costs and therefore reduces productivity. While that might be a factor, he cites weak investment limiting the ability of labor to leverage its efforts. Blinder further questions whether Twitter and other Internet services that people use at work actually are a form of leisure or entertainment and detract from productivity.
The Goldman Sachs report on productivity looks at changes in productivity over the last fifty years. It views the period from 1960 to 1970 as “the Golden Age” following IBM’s introduction of a computer capable of performing a variety of functions and Intel’s introduction of the microprocessor. Annual productivity in that decade rose to 3.5%, but it sagged in the 1970s, and in 1987, economist Robert Solow said, “Computers are everywhere but in the productivity figures.” In 1992, however, the worldwide web became available to the general public and by 2004 productivity was back up to 4%. Now, with 51% of all U.S. adults banking online and 90% having smart phones, it has dropped to 1% in 2014. Goldman suggests there are two areas of possible significant productivity growth. The first is the extractive industries of energy and mining. The second is so-called “old economy” industrial companies which still have a long way to go in implementing technology. Eric Schmidt of Google expects artificial intelligence to increase productivity in healthcare and the legal profession. “Any repetitive task can be done effectively by a computer,” he says.
If the low productivity numbers are real and not a result of measurement error, they are significant. If productivity does not improve, the implications for the future economic outlook are worrisome. The productivity report for the third quarter showed that employee output per hour increased at a 1.6% annual rate, well above the estimates of most economists. Productivity only increased .4% in the third quarter of 2014, but from 2000 to 2013 productivity improved 2.1% annually and that includes the recession of 2008-9. One explanation for the expectation of lower productivity gains is that there have been no significant technology breakthroughs in the last several years, as Robert Gordon pointed out.
The major contributor to the favorable third quarter productivity number was a decrease in the number of hours worked which dropped by the most in six years as a result of an 18% decline in the number of people working for themselves and not picked up in the surveys. There has been a lot of interest in the importance of the “gig economy” where people get paid for performing tasks but do not have a formal job. Many younger people like the freedom and independence of this work and are willing to give up the structure and benefits of traditional employment to engage in work that conforms to their desired lifestyle. It is hard to assess the long-term impact this phenomenon will have on the productivity data.
The equity market is very much aware of the benefits of producing profits with relatively few employees. Some of the best performing stocks have a very high enterprise value to employee ratio. According to Strategas Research, examples include Facebook with an enterprise value (EV) of $285 billion and 9,200 employees, for a ratio of 31, Visa with an EV value of $185 billion and 9,500 employees and a ratio of 20 and Netflix with an EV of $46 billion, 2,450 employees and a ratio of 19. We have transitioned to a knowledge-based economy where the computer plays an important role from a manufacturing economy where labor was key to output. There are two important implications of that change. The first is that an entrepreneur can create a very valuable company in a relatively short period of time; the second is that some of the most exciting companies get the job done with relatively few workers. In some cases they do so while displacing the work done by competitors who employ many more people. Compare the companies above with more worker-intensive examples like Gap with an EV value of $11 billion and 141,000 employees for a ratio of .08, and Yum! with an EV of $34 billion and 537,000 employees for a ratio of .06.
In terms of productivity, assessing the impact of technology is hard. There is no question that the cell phone and the computer have made us more productive. Many of us can work effectively away from the office in any location. There is a question as to whether time spent on Facebook is productive, entertaining or a distraction. In any case some of the new, innovative companies like Uber and Airbnb have productivity characteristics that are probably not accurately reflected in the reported statistics. What we do know is that thousands of jobs in manufacturing and services have been eliminated by technology. This has resulted in favorable productivity figures over the last fifty years and sent profit margins to an all-time high, allowed the stock market to recover and increased the perception of inequality. Both corporate profitability and the standard of living are tied to productivity. If productivity is being properly measured and is, in fact slowing, it will have a profound impact on the future outlook for the economy and the financial markets.