September Newsletter: The Productivity Paradox and Why it Matters to Investors

Posted by Alex Frey (@alexhfrey )

In George Orwell’s timeless book 1984, “doublethink” is the classic name given to the ability to simultaneously believe two contradictory things, a critical trait needed to psychologically manage the books “Orwellian” times. The actual date 1984 long ago came and went, but doublethink remains as relevant a metaphor as ever for describing how “collective wisdom” can sometimes spiral around itself. 

Take what we have previously called one of the three most significant questions for the future of the world economy: what is happening with the pace of technology improvement and productivity? Is it constantly accelerating, or slowing to a crawl? 

The collective wisdom seems to be: both.

On the one hand, transformative advances in technology are held to blame (along, occasionally, with Chinese workers and immigrants of all stripes) for killing off stable blue-collar manufacturing jobs. On the other hand, the absence of any good investment opportunities and technological morass are held up as reasons for low interest rates, anemic economic growth, and faltering productivity. 

We first touched on this issue about a year ago in a newsletter on trends in productivity growth, which is closely related to technological advances. As a refresher, productivity is the output of the economy, as measured by gross domestic product (GDP) divided by the number of hours worked; i.e., it is the amount which an average worker produces within an hour of labor. The puzzle we discussed is that despite rapid advances in computer technology, productivity (as reported by the Bureau of Labor Statistics) is currently growing at a lower rate than it did in most of the post-World War II era.   

Things have only gotten worse in the last year. Figure 1 is an updated version of a chart presented in the mentioned newsletter. It shows a sharp slow-down in productivity growth in the years from 2007 to 2015. Some economists believe that this is going to be the pattern of the future. Robert Gordon (reference 1), for instance, believes that the current rate of productivity growth is the norm and that earlier periods of higher productivity growth were associated with the exploitation of electrification and the internal combustion engine and that these high levels are not likely to be repeated.

Other economists, in particular Brynjolfsson and McAfee (references 2 and 3) believe we are on the cusp of a revolution where computers will replace many jobs and dramatically change the economy. Others in this camp believe that the start of these technological advances are a reason why employment and wages have both been slow to grow following the end of the Great Recession. 

So which is it: Is productivity on the cusp of increasing so quickly that computers will eventually do everything that we do, only better, causing widespread unemployment? Or are we stuck in a productivity “stagnation”, holding down economic growth for the foreseeable future?


Figure 1: The productivity growth rate for different time periods after World War II. 

First let’s review some basics. Society as a whole cannot consume more than it produces, so economists have always assumed that productivity growth is good for society. The general public has not always accepted this. Ever since the beginning of the Industrial Revolution in 18th century England, some people have worried that technology was going to take away their jobs, and some have tried to resist technical advances. In the early 19th century, Luddites roamed about in England trying to destroy the new mechanized spinning jennies and looms. In a more recent example, in the 1960s, unions rather needlessly tried to keep locomotive firemen in railroad locomotives long after trains had switched to diesel fuel, even though they would have no real function. 

To date, these efforts to impede technology have failed, and there has been a constant parade of jobs eliminated or vastly reduced in number by technology. Despite this, employment has not decreased, and the general standard of living has increased dramatically. When technology has replaced jobs, the cost of the goods produced has decreased, and this has liberated consumers to spend more on other things. That in turn creates new jobs that balance out those that have been lost, and everyone has a higher standard of living. This process has been going on continuously for at least 250 years. In the past, technology growth has clearly been a boon to society as a whole. 

Is this still true? Despite the Bureau of Labor Statistics (BLS) data, Brynjolfsson and McAfee (B&N) now believe that technology may be destroying jobs so rapidly that the economy can’t keep up. By this analysis, new jobs are not being created fast enough to replace those that are lost to technology. B&N have drawn attention to what they call the “Great Decoupling”. For most of the time since World War II, productivity and employment have grown in tandem, but starting in about 1990, employment decoupled from productivity and began to grow much slower than productivity. Figure 2 illustrates this point, and this is what B&N call the “Great Decoupling”. They also note that median income level has been stagnate since the late 1990s even though the economy has grown and the average income has increased significantly. This is illustrated by figure 3. They attribute these changes, at least in part, to the fact that technology is destroying middle income jobs faster than new jobs can be created.


Figure 2: The growth of productivity and employment in the post-World War II era. Taken from Brynjolfsson and McAfee, writing in the New York Times on December 11, 2012. 


Figure 3: Trends in productivity, GDP, employment, and median family income. Taken from Brynjolfsson and McAfee as presented in the Harvard Business Review, June, 2015. 

So how can we reconcile the views of B&N with the underlying statistics that indicate a productivity slowdown? One possibility we shouldn’t immediately dismiss is that those underlying statistics are wrong. Coming up with a “number” that represents the “size” of the overall economy was always a somewhat nebulous task, and the current approach that is used today (GDP) was designed for an industrial economy that focused on producing ever-more tangible things. 

There are many people who believe that the economy is changing in ways that render the old statistical measures obsolete. GDP is a measure of everything that is sold in the economy. Many things that we used to pay for are now free, and thus are not included in GDP. Google provides email and search services to consumers for free. Email reduces the need for postal workers; free search services eliminate the need for many travel agents, airline ticket agents, and others who used to be paid to share their knowledge with consumers. But, since they’re free, these new services don’t directly add anything to GDP (of course, Google sells ads, and this is reflected in GDP). Put differently, if today’s economy is increasingly focused on the “production” of intangible goods, GDP might not be the appropriate measure. 

While many of the methodological arguments of the technological optimists camp seem highly credible, there is also a strong case to be made that while there are strong advances in some sectors of the economy (manufacturing and IT-related), the parts where the most people work (health care, personal services, law, taxes, etc.) are not advancing at the same rate. Hair dressers, for instance, are probably not able to cut twice as many hairs per hour as they were 50 years ago, while that is about the improvement in underlying microprocessor technology every 18 months. So without a change, as we become more and more an economy of hairdressers and lawyers rather than factory workers, perhaps productivity will naturally tend to slow.

But assuming that productivity is increasing faster than the statistics show, does this mean that workers are indeed doomed? There are reasons to be both optimistic and pessimistic. 

On the optimistic side, the money that consumers save by having free or much cheaper services may be spent elsewhere to create new jobs. People that would have spent $800 on a digital camera instead use the phone on their new iphone 7 and spent the difference on an Amazon Echo and some new Bluetooth speakers. This is essentially what has happened in the past, and there is strong logic that “human demand is insatiable” and will never run out. 

One reason noted by B&N to be pessimistic is that increases in GDP are flowing disproportionately to the very rich, who may have little need or desire to spend it, and that may inhibit job formation. The tendency to increased inequality is a world-wide phenomenon, and it may be linked to technology. Digital technology can easily be transported around the world, and it tends to create “winner take all” products and companies. Whereas there were once many local markets where different providers could compete, for many products there is now only one world-wide market where one or two companies dominant the market. In a “winner take all” economy, being second best can equate to failure. The winners make large profits, the second best flounder. If B&N are correct, the rapid advance in technology may be a problem for society as a whole. Society as a whole will suffer unless the economic benefits of technology flow down to middle income and lower income people.
The effect on investors is not quite so obvious (we made some comments about this in the earlier newsletter). The effect of productivity growth on stock prices can be obscured by other factors, and studies of the relation between the two have not given consistent results. Rapidly advancing technology can be very bad for companies whose products are being supplanted (for instance, movies theaters, the Encyclopedia Britannica, and the Post Office). But of course, great profits can be made in the new businesses that replace the old. We assume that profits are likely to grow along with the economy, so there should be some positive correlation between productivity growth and stock prices. However, growing inequality might derail this process by stifling overall economic growth. If the trend of increasing inequality can be arrested or reversed, we believe that growing productivity will have positive effects for both investors and society as a whole.

Given that this is so, how should it affect your choice of investments? If technology will be driving future growth, one obvious strategy is to buy mutual funds or ETFs that specialize in technology stocks. One problem with this strategy is that these funds tend to have high price earnings (PE) ratios. As we write this, the PE on the Vanguard Information Technology Index Fund is 25.3, not an outlandish value, but not cheap either. The T. Rowe Price Sciences and Technology Fund has a PR ratio of 32.1. While these might be good buys for a portion of your portfolio, one should remember that low PE stocks (value stocks) have out-performed high PE stocks in the past, precisely because investors have tended to overestimate the impact of shiny new things. 

A second problem with investing directly in technology shares is that many of the firms that will produce the breakthrough technological advances of the next thirty years have likely not gone public, or in many cases even been formed yet. Just as investing in a technology ETF ten years ago would have given you zero exposure to Facebook, Twitter, or Uber, which are today three of the largest technology companies in the US, so too will investing in that ETF today likely not give you much ownership of tomorrow’s biggest winners. 

Another strategy would be to buy stocks in companies that produce things that are not likely to be replaced by technology and that will benefit from general economic growth. One problem is identifying what this is. Natural resource stocks are one example that people have been given in the past, since we will need energy and raw materials to power whatever we do. However even here it is not clear that we might not find new sources of energy and better materials to use in the future. For instance, advances in solar, wind, or even fusion power combined with advances in battery technology could dramatically undercut the value of the existing coal and oil reserves that many energy companies have.

Anyone planning to buy a mutual fund or ETF that specializes in natural resources should be aware that many of these funds are very heavily weighted towards energy. Two that have a broader base are the T. Rowe Price New Era Fund and the SPDR S&P Global Natural Resources ETF. These funds have a broad natural resource focus and currently have PE ratios that are slightly below the ratio for the Vanguard Total Stock Market Index Fund. Natural resource stocks have done poorly in the last few years, but this could possibly be a good time to buy in a “sandbox” or “play money” account (less than 20% of your overall portfolio is good rule of thumb).

So after all that, which is it? Should we fear too much productivity growth, too little productivity growth, or we will we end up in a Goldilocks state of the perfect amount? Our nuanced view is: bring on the productivity growth, but we will have to be on guard to see how the fruits of it are distributed (we’ll leave the political commentary to others – but our interest here is investing).

But more than driving at a conclusion, we find following these kinds of debates most illustrative for how they indicate the limits of our own human understanding of highly complex systems like the financial markets. The mere fact that smart people are debating whether technology is killing all of our jobs or actually stagnating tells you that we are all somewhat clueless on what is going on.

Respecting the limits of our imperfect knowledge is one big reason to follow a rules-based and risk-controlled system like the IvyVest model. 

1. Gordon, Robert J., “Revisiting U. S. Productivity Research Over the Last Century with a View to the Future”, National Bureau of Economic Research Working Paper Series, Paper No 15834, 2010.
2. Erik Brynjolfsson and Andrew McAfee, Race Against the Machine: How the Digital Revolution is Accelerating Innovation, Driving Productivity, and Irreversibly Transforming Employment and the Economy, Digital Frontier Press, Lexington, MA, 2011.
3. Erik Brynjolfsson and Andrew McAfee, The Second Machine Age: Work, Progress, and Prosperity in a Time of Brilliant Technologies, W.W. Norton Company, 2014

Get our next article delivered to your inbox.

Sign up below and be the first to know about our freshest data-driven thinking on the markets, and investing. We will send you no more than one email a week. This is free.

Ready to start putting this into action?

Take a free two-week trial to IvyVest premium -- our premium subscription service. You'll get access to our rules-based dynamic asset allocation model, tools that will show you exactly what you need to buy in your own discount brokerage account (and when to re-balance) to implement it for yourself, and an insightful monthly newsletter that will keep you on abreast of the most important things going on in the markets. There is no credit card required. Get Started Now!

By Alex Frey