Labor Costs, Inflation Expectations, and the Affordable Care Act: What Businesses Are Telling Us

The Atlanta Fed’s May survey of businesses showed little overall concern about near-term inflation. Year-ahead unit cost expectations averaged 2 percent, down a tenth from April and on par with business inflation expectations at this time last year.

OK, we’re going to guess this observation doesn’t exactly knock you off your chair. But here’s something we’ve been keeping an eye on that you might find interesting. When we ask firms about what role, if any, labor costs are likely to play in their prices over the next 12 months, an increasing proportion have been telling us they see a potential for upward price pressure coming from labor costs (see the chart).

(enlarge)

To investigate further, we posed a special question to our Business Inflation Expectations (BIE) panel regarding their expectations for compensation growth over the next 12 months: “Projecting ahead over the next 12 months, by roughly what percentage do you expect your firm’s average compensation per worker (including benefits) to change?”

We got a pretty large range of responses, but on average, firms told us they expect average compensation growth—including benefits—of 2.8 percent. That’s about a percent higher than the average over the past year (as estimated by either the index of compensation per hour or the employment cost index). But a 2.8 percent rise is also about a percentage point below average compensation growth before the recession. We’re included to read the survey as a confirmation that labor markets are improving and expected to improve further over the coming year. But we’re not inclined to interpret the survey data as an indication that the labor market is nearing full employment.

We’ve also been hearing more lately about the potential for the Affordable Care Act (ACA) to have a significant influence on labor costs and, presumably, to provide some upward price pressure. Indeed, several of our panelists commented on their concern about the influence of the ACA when they completed their May BIE survey. So can we tie any of this expected compensation growth to the ACA, a significant share of which is scheduled to go into effect eight months from now?

Because a disproportionate impact from the ACA will fall on firms that employ 50 or more workers, we separated our panel into firms with 50 or more employees, and those employing fewer than 50 workers. What we see is that average expected compensation growth is the same for the bigger employers and smaller employers. Moreover, the big firms in our sample report the same inflation expectation as the smaller firms.

But the data reveal that the bigger firms are a little more uncertain about their unit cost projections for the year ahead. OK, it’s not a big difference, but it is statistically significant. So while their cost and compensation expectations are not yet being affected by the prospect of the ACA, the act might be influencing their uncertainty about those potential costs.

(enlarge)

By Mike Bryan, vice president and senior economist,

Brent Meyer, economist, and

Nicholas Parker, senior economic research analyst, all in the Atlanta Fed’s research department

Behavior’s Place in the Labor Force Participation Rate Debate

It's not often that the mainstream media is interested in the nuances of labor market statistics, so last week’s debate over the meaning of labor force participation rates (LFPR) in the pages of the Washington Post and the Wall Street Journal was music to this labor economist's ears.

Sparked by an article by Ben Casselman in his April 29 Wall Street Journal Outlook column, the ensuing back and forth (here, here, here, and here) between Casselman and the Post’s Jim Tankersley focused on what has become a central preoccupation in assessing the likely course of the labor market: Is the recent decline in the labor force participation rate the result of structural factors (e.g., an aging population) or cyclical ones (such as weak economic conditions)? Almost contemporaneously, Bill McBride declared in his recent Calculated Risk blog, "…most of the [recent] decline in the participation rate was due to changing demographics...as opposed to economic weakness."

The changing pattern of labor force participation has been a topic of discussion among economists for some time—for example, see my Federal Reserve Bank of Atlanta Economic Review article—and both Tankersley and Casselman agree that the long-run secular decline in participation is a matter worthy of independent concern. But the Federal Open Market Committee has substantially raised the stakes on disentangling longer-run trends from short-run cyclical (and presumably temporary) movements in labor force participation. It’s done this by introducing into its policy deliberations concepts like unemployment thresholds and qualitative assessments on “substantial” labor market improvement.

Casselman, in an October 2012 WSJ article, cites work by my colleagues at the Chicago Fed, who find that while more than two-thirds of the decline in LFPR between 1999 and 2011 is accounted for by changes in the age distribution of the population, "…over the 2008-2011 period...only one-quarter of the...decline of actual LFPR...can be attributed to demographic factors."

This conclusion—that three-quarters of the decline in the LFPR since the beginning of the Great Recession can be attributed to cyclical factors—is supported by other research. Colleagues at the Kansas City Fed and at the Board of Governors concur that the vast majority of the decline in the LFPR since 2008 is the result of cyclical factors. Even economists outside the Federal Reserve System acknowledge the significant role of cyclical factors in the LFPR decline (for example, see the analysis by economists at the Deutsche Bank).

But there is a critical third piece to the LFPR puzzle that most of these studies ignore. In addition to changing demographics (which have, for example, been associated with a rising share of retirement-age individuals in the total population) and cyclical effects (for example, the tendency for participation to fall when wage growth is tepid or job opportunities scarce), there are also behavioral changes afoot—a point Casselman makes in his final installment of the Post/WSJ debate. For example, individuals of near-retirement age may extend their participation as a result of significant, unexpected declines in wealth. Or women with young children—a demographic group typically less likely to participate in the labor market—may increase participation if a partner loses a job during an economic downturn. In both cases, participation rates for these demographic groups would not fall by as much as expected in response to high unemployment rates alone.

Work that I've done with Fernando Rios-Avila, a colleague at Georgia State University, finds that more than 100 percent of the fall in the LFPR since 2008 is accounted for by the condition of the labor market (cyclical factors), but these particularly strong cyclical forces were countered by increased tendencies to participate (behavioral changes). In other words, if individuals hadn't stepped up to the plate and exhibited even stronger labor force participation behavior than before the recession, the LFPR would be even lower than it is.

To illustrate the role that changing behavior played in the LFPR decline during the Great Recession, the chart below illustrates how this decline can be separated into a trend component (demographics), a cyclical component (strength of the labor market), and a behavioral component. The solid black line reflects the actual LFPR in March of each year calculated using the Current Population Survey, which is the survey data used by the U.S. Bureau of Labor Statistics to calculate the monthly labor force statistics. The orange line reflects the trend estimate of the LFPR using only demographic data (such as the age distribution of the population) through 2007, projecting out to 2012. As many others have pointed out, changing demographics—the aging of the baby-boom generations, if you will—explains only about 30 percent (in this example) of the actual post-2007 decline in LFPR.

(enlarge)

But the chart also reveals something that may be underappreciated. Including a measure of labor market conditions in the projection of the LFPR, as well as a depiction of prerecession behavior (the green line), indicates that the LFPR should be much lower than it actually is. The message from this exercise is that the actual LFPR in 2012 was above what would have been projected had each demographic group exhibited the same labor force participation behavior after the recession as before the recession.

As it turns out, women, ethnic minorities, older people, and individuals with small children were much more likely to participate in the labor market after the recession than before it. These workers are often referred to as "added workers," or workers who join the labor force to make up for lost income elsewhere in the household. As I noted above, if these demographic groups had not increased their participation in the labor force, the aggregate LFPR would be much lower than it is.

What the chart tells us is that the cyclical factors affecting labor force participation are even more important than generally imagined. However, it is also true that the inevitable march of time will continue to put powerful downward pressure on labor force participation. Indeed, our research predicts only a modest rise in the LFPR if labor markets rebound to prerecession conditions. Our results suggest that relative to the the average LFPR over the years 2010–12, the average LFPR over the years 2015–17 will rise by about a third of a percentage point—again, if the labor market returns to prerecession conditions. Though higher than today, this level would still leave the LFPR considerably lower than it was before the recession, primarily reflecting the continued downward pressures of aging baby boomers.

By Julie Hotchkiss, a research economist and policy adviser in the Atlanta Fed’s research department

Weighing In on the Recent Discrepancy in the Inflation Statistics

Recently, there has been a divergence between inflation as measured by the Consumer Price Index (CPI) and the preferred inflation measure of the Federal Open Market Committee (FOMC), which is the price index for personal consumption expenditures (PCE). That divergence is fairly evident in the “core” measures of these two price statistics shown in the chart below.

(enlarge)

This strikes us (and others, like Reuters’ Pedro da Costa) as a pretty significant development. The core CPI is telling us that the underlying inflation trend is still holding reasonably close to the FOMC’s longer-term target of 2 percent. But the behavior of the core PCE is rather reminiscent of 2010, when the inflation statistics slid to uncomfortably low levels—a contributing factor to the FOMC’s adoption of QE2. Which of these inflation statistics are we to believe?

Part of the divergence between the two inflation measures is due to rents. Rents are rising at a good pace right now, and since it’s pretty clear that the CPI over-weights their influence, we might be inclined to dismiss some part of the CPI’s more elevated signal. But then there are all those “non-market” components that have been pulling the PCE inflation measure lower—and these aren’t in the CPI. These are components of the PCE price index for which there are no clearly observable transaction prices. They include the “cost” of services provided to households by nonprofit organizations, or the benefits households receive that can only be imputed (i.e., that “free” checking account your bank provides if you maintain a high balance.) Since we can’t really observe the price of these things, we’d probably be inclined to dismiss their influence on PCE the inflation measure. But we’ve done the math, and the impact of these two influences accounts for only about a third of the recent gap between the core PCE and the core CPI inflation measures. Most of the disagreement between the two inflation estimates is coming from elsewhere.

We could continue to parse, item by item, all the various components and weights of the two statistics to get to the bottom of this discrepancy. But in the end, such an accounting exercise would merely tell us why the gap between the two measures has emerged, not which measure is giving the best signal of emerging inflation trends.

As an alternative approach, we thought we’d let the data speak for themselves and search for a common trend that runs through the detailed price data. What we have in mind is to compute the “first principal component” of the disaggregated data used to calculate the CPI and the PCE price indexes. The first principal component is a weighting of the data that explains as much of the data variation as possible. So, in effect, the detailed price data in each price index are being reweighted in a way that reveals their most commonly shared trend, and not by their share of consumer expenditure.

The chart below shows the 12-month trend of the first principal component derived from the 45 CPI components used in the computation of the Federal Reserve Bank of Cleveland’s median CPI, and the first principal component derived from the 177 components used in the computation of the Federal Reserve Bank of Dallas’s trimmed-mean PCE. (These are the most detailed component price data we could easily get our hands on.)

(enlarge)

So what do we make of this picture? Well, three things:

First, inflation as measured by the PCE price index has tended to track about 0.25 percentage point under inflation as measured by the CPI over time. So part of the gap between the two inflation measures appears to be a long-term feature of the two inflation statistics.

Second, the first principal components of both the CPI and the PCE data have been persistently under their precrisis averages. In the case of the PCE measure, the first principal component is under the FOMC’s 2 percent target (a point that has not gone unnoticed by Paul Krugman).

A third takeaway from the chart is that the “disinflation” pattern traced out by these principal components has been gradual and modest—much more so than what the core PCE has recently indicated and what the data were telling us back in 2010.

Does that mean we should ignore the recent disinflation being exhibited in the core PCE inflation measure? Well, let’s put it this way: If you’re a glass-half-full sort, we’d say that the recent disinflation trend exhibited by the PCE price index doesn’t seem to be “woven” into the detailed price data, and it certainly doesn’t look like what we saw in 2010. But to you glass-half-empty types, we’d also point out that getting the inflation trend up to 2 percent is proving to be a curiously difficult task.

By Mike Bryan, vice president and senior economist,

Pat Higgins, economist,

Brent Meyer, economist, and

Nicholas Parker, senior economic research analyst, all in the Atlanta Fed’s research department

Improvement in the Outlook? The BIE Panel Thinks So

Earlier this month, Dennis Lockhart, the Atlanta Fed’s top guy, gave his assessment of the economy and monetary policy to the Kiwanis Club of Birmingham, Alabama. Here’s the essential takeaway:

There are encouraging developments in the economy, to be sure, but the evidence of sustainable momentum that will deliver “substantial improvement in the outlook for the labor market” is not yet conclusive. ... How will I, as one policymaker, determine that the balance has shifted and the time for a policy change has come? Well, one key consideration is the array of risks to the economic outlook and my degree of confidence in the outlook.

To help the boss assess the risks to the outlook, we reached out to our Business Inflation Expectations (BIE) panel to get a sense of how they view the outlook for their businesses and, notably, how they assess the risks to that outlook. Specifically, we asked:

Projecting ahead, to the best of your ability, please assign a percent likelihood to the following changes to UNIT SALES LEVELS over the next 12 months.

The table below summarizes the answers and compares them to the responses we got to this statement last November.

(enlarge)

First, the business outlook of our panel has improved decidedly since last November. On average, our panel sees unit sales growth averaging 1.8 percent. OK, not a spectacular number, but, to our eyes at least, much improved from the 1.2 percent the group was expecting when we queried five months ago.

And how about the assessment of the risks President Lockhart indicated was also a key consideration? Here again, the sentiment in our panel appears to have shifted favorably. Last November, our panel put the likelihood that their year-ahead unit sales growth would be 1 percent or less at 50 percent. The group now puts the chances of a downshift in business activity at 37 percent. Meanwhile, the upside potential for their sales has grown. Last November, the panel put the chances of a “significant” improvement in unit sales at about 20 percent; this month, the group thinks the likelihood is 30 percent.

And this improved sentiment isn’t centered in just a few industries—it’s spread across a wide swath of the economy. Firms in construction and real estate, which were, on average, projecting 12-month unit sales growth of 1.1 percent last November, now put that growth number at 1.8 percent. The average sales outlook of general-services firms has risen from 1 percent to 2.2 percent; finance and insurance companies went from 0.5 percent to 1.3 percent; and retailers/wholesalers’ unit sales projections rose from 1.5 percent to 2 percent. And manufacturers, who posted relatively strong expectations last November, reported about the same sales outlook this month as they did five months ago.

(enlarge)

To be clear, President Lockhart’s recent comments—and the Federal Open Market Committee statement on which they are based—indicate he is looking for a substantial improvement in the outlook for the labor market, not sales. But we’re going to assume that it’s unlikely to have one without the having the other. And is our panel’s unit sales forecast “substantially” improved? Well, what constitutes “substantial” is in the eye of the beholder, but if this isn’t a substantial improvement in the outlook, it’s certainly a move in that direction.

By Mike Bryan, vice president and senior economist, and

Nick Parker, economic research analyst, both in the Atlanta Fed’s research department

Higher Education: A Deflating Bubble?

There are at least two sides to every debate, but it’s becoming clearer by the day that the debate over the cost of higher education is being won by people like University of Tennessee law professor Glenn Reynolds.

A frequent writer and lecturer, and even more frequent blogger, Reynolds visited the Atlanta Fed recently to share his views with local community leaders. He reported that total student loan debt now stands at over $1 trillion—more than total credit card debt and auto loan debt combined. As these charts from the New York Fed show, the increase in total student debt over the past eight years is a result of greater numbers of students and families taking on educational debt as well as higher debt balances per student. 

One can argue that this trend is not necessarily a bad thing. Education is an investment in human capital, and if those newly acquired skills are valued highly by employers, then going to college can be a positive net present value project, even with debt financing.

And wage data reveal that these skills are indeed valuable. As this Cleveland Fed article and chart show, the median wage for a worker with a bachelor’s degree was about 30 percent higher than that of a worker with only a high school diploma in the late 1970s and grew to more than 60 percent higher by the early 2000s. However, the data also show that over the last decade the value of a college degree measured by wages has stagnated.

And here begins the crux of Reynolds’s concern. The cost of attending college has continued to grow, and grow rapidly. Between the 2000–01 and 2010–11 academic school years, the cost of undergraduate tuition, room, and board rose 42 percent at public institutions and 31 percent at private not-for-profit institutions, after adjusting for inflation, according to the National Center for Education Statistics.

A stagnant wage premium with rising costs of attendance suggests that, at least on average, the value proposition of going to college is deteriorating. To make matters worse, Reynolds described students graduating with significant levels of student loan debt who often cannot find jobs that pay enough to cover the loan payments. Moreover, unlike credit card debt, student loans are not dischargeable in bankruptcy, meaning that there is no opportunity to get out from under the debt burden other than through full repayment. Reynolds told of individuals whose high levels of student debt are limiting their career choices, ability to obtain mortgages, and save for retirement. He even went on to say that student loans are affecting a much more personal market—the marriage market. After all, he says, “Who wants to marry someone with huge amounts of unpayable debt?”

Reynolds contends that ”something that can’t go on forever, won’t,” and he believes that seeing friends or family members having financial problems because of student loans is leading college students to become more cost conscious. Additionally, he notes that more and more of today’s students are focusing on majors that seem likely to offer a strong salary over time. The chart on 2009 enrollment and wage premiums by major show some support for that notion.

Large fractions of students are enrolled in majors with relatively higher wage premiums, including business and engineering, but there are also substantial enrollments in education, psychology, and the humanities. For Reynolds it is not so much about seeking out the highest-wage major; instead, his advice is, “Don’t go to a college that will require you to borrow a lot of money.”

What’s the endgame? Well, he expects that when the bubble bursts, there will be less “dumb money” to be gained, students will demand a higher return on investment, and schools will ultimately be forced to adapt. According to Reynolds, colleges have two different strategic choices: increase the value of the education for the current cost, or lower the cost of providing the current level of value. And he expects the most common response will be the latter, likely involving technology such as MOOCs (massive open online courses) and other innovations in teaching methods.

When any bubble bursts, there are some casualties. In this case, it may be that some colleges do not survive once market discipline has been unleashed. Given the statistics above, you might think that it would be the small liberal arts colleges that will suffer the most, but in this video, shot during the visit to Atlanta, Reynolds argues that these colleges may actually gain from the coming shakeout.

Reynolds indicated that there is change in the air, but it’s coming slowly. The bubble may not have burst, but he sees it deflating. He noted, “A lot of people hope it will pass. They’ll muddle through without dramatic changes. And frankly I hope they’re right. But I don’t think they are.”

By Paula Tkac, vice president and senior economist in the Atlanta Fed research department and

Michael Chriszt, vice president and community relations officer in the Atlanta Fed’s public affairs department

Being Ahead of the Curve: Not Always a Good Thing

Our friends at the New York Fed have a nifty interactive graphic that compares the unemployment rate, labor force participation rate, and employment-to-population ratio over the last five business cycles. You can even break these indicators down by gender, by age, or by a particular business cycle. (For a deeper dive, check out this post at Liberty Street Economics by Jonathan McCarthy and Simon Potter.) And though it’s not exactly late-breaking news, no matter which of the three indicators you look at, you can’t help but conclude that the most recent recession is an outlier.

The Beveridge curve is a fourth and particularly useful graphical representation of a steady-state economy showing how, in theory, one might expect the vacancy rate to change, given an unemployment rate. It depicts the relationship between job openings and the unemployment rate. (The Atlanta Fed’s magazine, EconSouth, discussed the Beveridge curve.) It, too, has been standing out over the course of the most recent recovery, so much so that we think it warrants at least a second glance. There are a number of ways to estimate a Beveridge curve (see, for example, methods described by Gadi Barlevy of the Chicago Fed here and by the Richmond Fed’s Thomas Lubik here).

We use the method described by Barnichon et al. (2012) to estimate the solid curve used in the first chart below. The square plots represent actual vacancy rate (y-axis) and unemployment rate (x-axis) combinations by month from December 2000, when the Job Openings and Labor Turnover Statistics (JOLTS) data series from the U.S. Bureau of Labor Statistics (BLS) series begins, to January 2013, the most recent month of data available for both series.

(enlarge)

Blue squares represent data from December 2000 to December 2009, when the “errors” between actual plots and the curve estimation were below 2 percentage points, and red squares represent data since January 2010, where data suddenly seem to jump higher than the predicted Beveridge curve to the tune of 2 percentage points or greater (see the chart below).

(enlarge)

But why?

In June 2012, Regis Barnichon and his coauthors concluded that the unemployment rate’s lackluster performance so far in the recovery was attributable to a shortfall in hires per vacancy. Since then, the vacancy rate has climbed its way back to its June 2008 level of 2.7 percent. However, the unemployment rate has clearly not returned to either its June 2008 level (5.6 percent) or where the Beveridge curve says it should be given this vacancy rate, which one might predict to be 5.5 percent using the methodology of Barnichon et al.

This “ahead of the curve” phenomenon has not gone unnoticed and has prompted some explanations. In a March 6, 2013, article in The New York Times (which also has some cool charts), Catherine Rampell posits that available positions are staying unfilled longer, while interview processes have become lengthier.

The next day, Rampell went into more detail about why we’re going “off the curve” in a New York Times Economix post. She cites skills mismatch and a skills atrophy effect of the long-term unemployed affecting the ability of employers to fill positions (which she explains aren’t full explanations, yet we would expect to see wages for highly coveted positions rise significantly).

Rampell goes back to the explanation many of us continue to hear from business contacts: employers are unwilling to fill vacant positions because of economic and fiscal policy uncertainty. She quotes Stephen Davis of Chicago’s Booth School: “They’re taking longer to fill vacancies because they just feel less need to fill jobs now,” Davis said. “They recognize that in a slack labor market, there is an abundance of viable candidates. If something happens, and if they need to hire quickly, they know they can do that. That’s harder in a tight labor market.”

So maybe as labor markets “tighten up,” or perhaps if the speed by which they tighten up quickens, we’ll get back on the Beveridge curve. Only time, and several BLS releases, will tell.

By Patrick Higgins, an economist at the Atlanta Fed, and



Mark Carter, a senior economic analyst at the Atlanta Fed

You Say You’re a Homeowner and Not a Renter? Think Again.

As we’ve said before, we’re suckers for cool charts. The latest that caught our eye is the following one, originally created by the U.S. Bureau of Labor Statistics (BLS). It highlights the relative importance assigned to the various components of the consumer price index (CPI) and shows where increases in the index have come from over the past 12 months.

It probably won’t surprise anyone that the drop in gasoline prices (found in the transportation component) exerted downward pressure on the CPI last year, while the cost of medical care pushed the price index higher. What might surprise you is the size of that big, blue square labeled “housing.” Housing accounts for a little more than 40 percent of the CPI market basket and, given its weight, any change in this component significantly affects the overall index.

This begs the question: In light of the recent strength seen in the housing market—and notably the nearly 10 percent rise in home prices over the past 12 months—are housing costs likely to exert more upward pressure on the CPI?

Before we dive into this question, it’s important to understand that home prices do not directly enter into the computation of the CPI (or the personal consumption expenditures [PCE] price index, for that matter). This is because a home is an asset, and an increase in its value does not impose a “cost” on the homeowner. But there is a cost that homeowners face in addition to home maintenance and utilities, and that’s the implied rent they incur by living in their home rather than renting it out. In effect, every homeowner is his or her own tenant, and the rent they forgo each month is called the “owners’ equivalent rent” (or OER) in the CPI. OER represents about 24 percent of the CPI (and about 11 percent of the PCE price index). The CPI captures this OER cost (sensibly, in our view) by measuring the cost of home rentals (details here). So whether the robust rise in home prices will influence the behavior of the CPI this year depends on whether rising home prices influence home rents.

So what is likely to happen to OER given the continued increase in home prices? Well, higher home prices, in time, ought to cause home rents to rise, putting upward pressure on the CPI. Homes are assets to landlords, after all, and landlords (like all investors) require an adequate return on their investments. Let’s call this the “asset market influence” of home prices on home rents. But the rents that landlords charge also compete with homeownership. If renters decide to become homeowners, the rental market loses customers, which should push home rents in the opposite direction of home prices for a time. Let’s call this the “substitution influence” on rent prices.

Consider the following charts, which show three-month home prices and home rents (measured by the CPI’s OER measure). It’s a little hard to see a clear correlation between these two measures.

So we’ve separated these data into their trend and cycle components (using Hodrick-Prescott procedures, if you must know) shown in the following two charts. Now, if one takes the trend view, there is a clear positive relationship between home prices and home rents. This is consistent with the asset market influence described above. But also consider the detrended perspective. Here, home prices and home rents are pretty clearly negatively correlated. This, to us, looks like the substitution influence described above.

So let’s get back to the question at hand. What do rising home prices mean for OER and, ultimately, the behavior of the CPI? Well, it’s rather hard to say because the link between home prices and OER isn’t particularly strong.

Not definitive enough for you? OK, how about this: We think the recent rise in home prices will more likely lean against the rise in OER for the near term as the growing demand for home ownership provides some competition to the rental market. But, in time, these influences will give way to the asset market fundamentals, and rents are likely to accelerate as returns on real estate investments are reaffirmed.

By Mike Bryan, vice president and senior economist, and



Nick Parker, economic research analyst, both in the Atlanta Fed’s research department

Half-Full Glasses

Just in case you were inclined to drop the "dismal" from the "dismal science," Northwestern University professor Robert Gordon has been doing his best to talk you out of it. His most recent dose of glumness was offered up in a recent Wall Street Journal article that repeats an argument he has been making for a while now:

The growth of the past century wasn't built on manna from heaven. It resulted in large part from a remarkable set of inventions between 1875 and 1900...

This narrow time frame saw the introduction of running water and indoor plumbing, the greatest event in the history of female liberation, as women were freed from carrying literally tons of water each year. The telephone, phonograph, motion picture and radio also sprang into existence. The period after World War II saw another great spurt of invention, with the development of television, air conditioning, the jet plane and the interstate highway system…

Innovation continues apace today, and many of those developing and funding new technologies recoil with disbelief at my suggestion that we have left behind the era of truly important changes in our standard of living…

Gordon goes on to explain why he thinks potential growth-enhancing developments such as advances in healthcare, leaps in energy-production technologies, and 3-D printing are just not up to late-19th-century snuff in their capacity to better the lot of the average citizen. To paraphrase, your great-granddaddy's inventions beat the stuffing out of yours.

There has been a lot of commentary about Professor Gordon's body of work—just a few examples from the blogosphere include Paul Krugman, John Cochrane, Free Exchange (at The Economist), Gary Becker, and Thomas Edsall (who includes commentary from a collection of first-rate economists). Most of these posts note the current-day maladies that Gordon offers up to furrow the brow of the growth optimists. Among these are the following:

And inequality in America will continue to grow, driven by poor educational outcomes at the bottom and the rewards of globalization at the top, as American CEOs reap the benefits of multinational sales to emerging markets. From 1993 to 2008, income growth among the bottom 99% of earners was 0.5 points slower than the economy's overall growth rate.

Serious considerations, to be sure, but there is actually a chance that some of the "headwinds" that Gordon emphasizes are signs that something really big is afoot. In fact, Gordon's headwinds remind me of this passage, from a paper by economists Jeremy Greenwood and Mehmet Yorukoglu published about 15 years ago:

A simple story is told here that connects the rate of technological progress to the level of income inequality and productivity growth. The idea is this. Imagine that a leap in the state of technology occurs and that this jump is incarnated in new machines, such as information technologies. Suppose that the adoption of new technologies involves a significant cost in terms of learning and that skilled labor has an advantage at learning. Then the advance in technology will be associated with an increase in the demand for skill needed to implement it. Hence the skill premium will rise and income inequality will widen. In the early phases the new technologies may not be operated efficiently due to a dearth of experience. Productivity growth may appear to stall as the economy undertakes the (unmeasured) investment in knowledge needed to get the new technologies running closer to their full potential. The coincidence of rapid technological change, widening inequality, and a slowdown in productivity growth is not without precedence in economic history.

Greenwood and Yorukoglu go on to assess, in detail, how durable-goods prices, inequality, and productivity actually behaved in the first and second industrial revolutions. They conclude that game-changing technologies have, in history, been initially associated with falling capital prices, rising inequality, and falling productivity. Here is a representative chart, depicting the period (which was rich with technological advance) leading up to Gordon's (undeniably) golden age:

Source: "1974," Jeremy Greenwood and Mehmet Yorukoglu,
Carnegie-Rochester Conference Series on Public Policy, 46, 1997

Greenwood and Yorukoglu conclude their study with this pointed question:

Plunging prices for new technologies, a surge in wage inequality, and a slump in the advance of labor productivity - could all this be the hallmark of the dawn of an industrial revolution? Just as the steam engine shook 18th-century England, and electricity rattled 19th-century America, are information technologies now rocking the 20th-century economy?

I don't know (and nobody knows) if the dark-before-the-dawn possibility described by Greenwood and Yorukoglu is the apt analogy for where the U.S. (and global) economy sits today. (Update: Clark Nardinelli also discussed this notion.) But I will bet you there was some commentator writing in 1870 who sounded an awful lot like Professor Gordon.

By Dave Altig, executive vice president and research director of the Atlanta Fed