Category Archives: disaggregating the economy

Macroeconomics I can approve

Posted on by

Raj Chetty and others write,

we study the mechanisms through which COVID-19 affected the economy by analyzing heterogeneity in its impacts. We first show that high-income individuals reduced spending sharply in mid-March 2020, particularly in areas with high rates of COVID-19 infection and in sectors that require in-person interaction. This reduction in spending greatly reduced the revenues of businesses that cater to high-income households in person, notably small businesses in affluent ZIP codes. These businesses laid off many of their employees, leading to widespread job losses especially among low-wage workers in affluent areas. High-wage workers experienced a “V-shaped” recession that lasted a few weeks in terms of employment loss, whereas low-wage workers experienced much larger job losses that persisted for several months. Building on this diagnostic analysis, we use event study designs to estimate the causal effects of policies aimed at mitigating the adverse impacts of COVID-19. State-ordered reopenings of economies have small impacts on spending and employment. Stimulus payments to low-income households increased consumer spending sharply, but little of this increased spending flowed to businesses most affected by the COVID-19 shock, dampening its impacts on employment. Paycheck Protection Program loans increased employment at small businesses by only 2%, implying a cost of $300,000 per job saved. These results suggest that traditional macroeconomic tools – stimulating aggregate demand or providing liquidity to businesses – have diminished capacity to restore employment when consumer spending is constrained by health concerns

Pointer from Tyler Cowen.

Aggregate economic data

Posted on by

For the essay I am writing on Edward Leamer, I have been re-reading Macroeconomic Patterns and Stories. He is really good at diving into data. Most academic macroeconomists think that actually studying the way that statistics are collected and looking for patterns would be beneath them.

For example, Leamer found the pattern of momentum in payroll employment growth. From that standpoint, the period from October 2011 through the September 2019 has been really uninteresting. There have been no sequences of three months of employment gains that were either well above or well below average. My rule of thumb is that less than 50,000 net jobs is a low-growth month and more than 350,000 is a high-growth month. Over the past eight years or so we have had only 3 low-growth months and one high-growth month (which barely made the cut at 355,000). So no real chance to test the momentum pattern, although I suppose that you count the persistence of middling numbers as “momentum.’

Anyway, I have the following thoughts, not all of which comes from Leamer.

1. The short-term fluctuations in GDP and net employment changes that we call recessions, even deep ones, are really small relative to: long-term growth; seasonal fluctuations (Leamer points out that real GDP on average drops at a 20 percent annual rate in the first quarter of the year, before the Commerce Department does its “seasonal adjustment” of the data. That is more than twice as much as the rate at which (seasonally-adjusted) real GDP falls during a bad recession.); secular shifts between sectors, e.g. employment rising in health care and education but declining in manufacturing; gross job flows, with 4 million workers leaving jobs and 4 million workers starting new jobs every month.

2. Can we say that the process for calculating real GDP in 2019 resembles the process in 2009? In 1999? In 1979? In 1959? In 1929? It would be an interesting exercise to go back to the raw data collected by the Commerce Department (or by Simon Kuznets prior to that) to estimate GDP. Fifty years ago, the government statisticians were calling manufacturing firms and getting counts of steel production or automobile assemblies or what have you. Now those figures are a much smaller fraction of GDP. How are the statisticians calculating the output of hospitals, of non-profit organizations, etc.? Academic economists don’t want to know how the statistical sausage gets made, but that seems to me to be a serious oversight.

3. Overall, I recommend being very wary of macroeconomic analysis that purports to give trends in productivity or to compare real income today with real income in past decades. The range of different answer that you can get using reasonable alternative methods for constructing the data far exceeds the variation in the phenomena that you are trying to assess.

What gets expensive and why

Posted on by

Eric Helland and Alex Tabarrok sort out the various proposed explanations. For example, concerning (lower) education, they write,

no metric of school quality shows any improvement that would appear to justify a cost increase of more than five times. Improvements in quality do not appear to explain increases in cost.

. . . Contrary to the usual story, the number of teachers per 100 students has increased since 1950. . . The number of other staff per 100 students also has increased, but at least since 1980 the increase has, if anything, been at a slower rate than the increase in teachers per student.

The rising cost of labor inputs is the best explanation for the rising cost of education

They focus on the Baumol Effect, about which I wrote

for everything that gets cheaper, something else has to get relatively more expensive. If efficiency shoots up in one sector, then in relative terms it has to decline elsewhere.

Divergence in state population trends

Posted on by

Antonio Chaves writes,

According to a study cited by the San Diego Union-Tribune, the majority of people who left California between 2007 and 2016 made less than $55,000 per year, and according to the New York Times, skyrocketing real estate prices have all but obliterated the black population in San Francisco. This black exodus is not limited to California. According to Forbes, many are also leaving Northern and Midwestern cities and moving to Sun Belt states in pursuit of better jobs and affordable housing.

The article assembles a lot of data on states that are gaining population and states that are losing population. The former tend to be red states and the latter tend to be blue states.

Disaggregated economy watch

Posted on by

Mark Perry updates his chart on divergent inflation trends.

Seven of those goods and services have increased more than average inflation, led by hospital services (+211%), college tuition (+183.8%), and college textbooks (+183.6%). Average wages have also increased more than average inflation since January 1998, by 80.2%, indicating an increase in real wages over the last several decades.

The other seven price series have declined since January 1998, led by TVs (-97%), toys (-74%), software (-68%) and cell phone service (-53%). The CPI series for new cars, household furnishings (furniture, appliances, window coverings, lamps, dishes, etc.) and clothing have remained relatively flat for the last 21 years while average prices have increased by 56% and wages increased 80.2%.

These cross-sectional differences in price movements are an order of magnitude greater than the time-series variation of “the” inflation rate. Indeed, I read these data as saying that “the” inflation rate is pretty close to a meaningless number

Colin Woodard watch

Posted on by

Adam Ramey writes,

we show that the migrations of millions of Okies from the central plains to California has a demonstrable effect on political outcomes to this day, even after accounting for other relevant geographic and demographic factors. After demonstrating this pattern at the electoral level, we leverage a decade’s worth of survey data and show that Hispanics living in areas with large Okie migrations in the 1930s are much more likely to have conservative social values and, importantly, to vote and identify as Republicans. Put together, these results suggest that the historical legacies of migration can have a strong and sustained impact even after nearly a century after the fact.

Pointer from Tyler Cowen.

Woodard has staked out the position that cultural differences across U.S. regions are the result of early settlement patterns. He does not hesitate to include in “Greater Appalachia” regions far from the vicinity of the mountain range. In the nineteenth century, folks migrated from Appalachia to parts of the Midwest and to rural Texas and Oklahoma. Then in the 1930s they settled in parts of California.

Productivity divergence

Posted on by

The WSJ reports,

According to data on advanced economies from the Organization for Economic Cooperation and Development, the most productive 5% of manufacturers increased their productivity by 33% between 2001 and 2013, while productivity leaders in services boosted theirs by 44%.

Over the same period, all other manufacturers managed to improve productivity by only 7%, while other service providers recorded only a 5% increase.

Think of a firm as consisting of labor, capital, and intangibles. The intangibles include knowledge and business strategy.

When intangibles hardly matter, then capital and labor ought to be about equally productive across all firms. When intangibles matter a lot, then productivity differences will widen.

Variation in occupational satisfaction

Posted on by

Greg Kaplan and Sam Schulhofer-Wohl write in an abstract,

The physical toll of work is smaller now than in 1950, with workers shifting away from occupations in which people report experiencing tiredness and pain. The emotional consequences of the changing occupation distribution vary substantially across demographic groups. Work has become happier and more meaningful for women, but more stressful and less meaningful for men. These changes appear to be concentrated at lower education levels.

I have said that this is worth studying.