Thursday, January 29, 2015

The Deflation Coin Flip

By my reckoning, there is now about a 50-50 chance the U.S. sees deflation in 2015, as measured by the year-over-year percentage change in the headline personal consumption expenditures price index, the measure favored by the Federal Reserve.

Above, you see my point estimates from a basic econometric model -- I've just taken headline and core PCE and the monthly average Brent crude spot and put them into a vector autoregression. The forecast was done on monthly log changes and then summed up to get year-over-year results. (This is an update of my earlier forecast here.)

In the next few months, I expect to see an abrupt disinflation, as lower oil prices feed into the price index. That will be followed by a few months of hovering at zero, with a bottoming in the summer. And then we will have a few months of inflation rising quickly back to normal.

That is contingent upon oil prices staying about where they are. Remember: If it's just a change in the level of oil prices, we will see a U-shape pattern in inflation, as we saw in 1986, 2008, and 2011. One implication of this is that we should not treat the decline in inflation as highly relevant for macroeconomic policy.

Another implication, though, is that it will be critical to keep a focus on some measure of core prices. Anticipating some pass-through from energy prices, any substantial move in core inflation would be unexpected and concerning, particularly in light of the recent decline in inflation breakevens.

Friday, January 9, 2015


I've just returned from a long trip to Malaysia as a member of Princeton's delegation to the World Universities Debating Championship, which Kuala Lumpur hosted this year.

I managed to take this panoramic shot of the city's skyline from Lake Titiwangsa, which was left behind by an abandoned open tin-mine pit and is now situated in a park to the north of the city center. A full-size copy of the image can be found here, and I release it into the public domain.

It was an amazing opportunity and great privilege to go, although perhaps not at the best time. The AirAsia flight went down two days after I arrived, and the country was in mourning. Beyond debate, I also learned an enormous amount as an economist -- just observing an East Asian developing economy firsthand was invaluable.

I'll try to put together some thoughts over the next few days if people are interested. As a start, I found this comparison helpful: Malaysia's GDP per capita, adjusted for PPP, is now about about a third of that of the US, making it comparable to Mexico today or South Korea in the mid-1980s.

Monday, December 29, 2014

Falling Participation, an Update

I recently updated my analysis of the decline in the labor force participation rate using data from the March 2014 Supplement to Current Population Survey.

The top-line result is that, of the 3.1 percentage-point decline in the participation rate between March 2007 and March 2014, 1.9 p.p (three-fifths) is explained demographic change and 1.2 p.p (two-fifths) of it is unexplained by demographic change. Of the explained portion, almost all of that (1.6 p.p.) is aging.

A note on interpreting these results: Econometric analyses of labor-market slack, I think, are helpful in getting a handle on what matters and what doesn't, and roughly how much certain factors matter. But don't hang on every decimal point. Take it into consideration along with the results of other analyses.

Sunday, October 12, 2014

Disinflation, Here We Come

Update (10/14)

It seemed worth bringing this question to the data. I built a simple three-variable vector autoregression of monthly PCE headline and core inflation and the monthly change in the Brent crude spot.

Assuming a price of $84/barrel -- yesterday's close -- I get that headline PCE inflation should bottom out at 0.9 percent in June 2015, incorporating in base effects. So the back-of-the-envelope calculations below check out nicely. We'll see if the model's forecasts hold up.

For future reference, the rule of thumb that comes out of the impulse response function is that for every 10-percentage-point increase in the spot price of Brent crude, expect 0.8 percentage points more in annualized headline PCE inflation for the next four months.

*     *     *

Spot prices for oil have dropped 20 percent in the last three months, from $110 to $90 a barrel. If they remain at these levels, inflation in the United States will slow quite a bit, and quickly at that. My estimate is that headline PCE inflation will fall to just under 1 percent within the next three months of data.

The estimate isn't that difficult to reach. Gas prices track closely with Brent crude, and some of the drop has already filtered into average prices at the pump, and some of that might already be in the August PCE data -- it's not clear exactly how much.

Suppose gasoline, as counted in PCE, end up 10 percent cheaper in September than in August. And energy goods and services is about 5 percent of personal consumption, of which 3 percentage points of that is directly gasoline. The other energy prices moves in tandem, and there are plenty of other prices in the economy that are sensitive to gasoline -- so that gives you about a half percentage-point drop in PCE inflation.

Recall that PCE inflation is 1.5 percent year-over-year. So, one surprise from energy markets, and we could be below one percent. At a time when we are supposed to be a couple months away from a rate hike, this could complicate the exit plan.

Saturday, September 27, 2014

What I Am Up To

Upperclassmen at Princeton must complete independent research, and Ilyana Kuziemko will be my adviser for junior year. I am still figuring out my research topic, but she focuses on inequality, criminal justice, public health insurance, and education -- so it is likely my topic will be in that general area. I usually do not share updates from my life at Princeton, but in case anyone has suggestions for things I should read or specific areas I might explore, I would very much appreciate it if you could reach out to me.

Update (10/29):

So I am doing research on that classic topic of labor economics: job training programs. It turns out that some questions about job training have been buried in the Basic CPS since 1994 to present and in the March Supplement from 2001 to 2009. I found them.

Using those data, I'll be examining the evolution of these programs since 1994, including the major 1998 reform, how the programs differ by type, and how the programs respond to business cycles.

Since so much work has been done estimating the treatment effects of these programs -- do they reduce unemployment? do they raise wages? -- I'm also going to do some work on another question using data from the American Time Use Survey: What's the elasticity of total search time with respect to participation? That is, do these programs substitute for other search activity, or does participation actually increase search time?

The goal here is to work towards answering the big question of how, not just if, job training programs work in the context of labor market search.

More on this soon.

Update (12/1):

It turns out there are insufficiently many observations of job training program participants in ATUS data (~300 pooled over 2003 to 2013). For what it's worth, which isn't much, my estimates of the elasticity of search time with respect to participation was about 2. That is, participation doubled search time -- depending on the kind of labor market search model in which this is contextualized, I imagine that is sufficient to explain all of the gains in labor market outcomes. Perhaps even more than sufficient, so that one would find that participation is effectively requiring the use of an inferior search technology. Interesting stuff, still, but one needs a bigger survey. If you find this question interesting and have the resources to make such a survey happen, please do reach out.

All this is to say I changed the topic of my research. Doing work now on the Supplemental Nutrition Assistance Program (SNAP, formerly the Food Stamp Program), specifically in the vein of Raj Chetty's 2008 paper on unemployment insurance, illiquidity, and moral hazard.

Monday, September 1, 2014

Tipped Workers and Hours

A couple summers ago, I worked as a busboy at a seafood restaurant on the Jersey Shore. The restaurant no longer exists. Hurricane Sandy wiped it off the map. But I was reminded of that summer on Labor Day.

Raising the federal minimum wage, now $7.25, has become a greater priority for public policy over the last couple years. My home state of New Jersey raised it by referendum to $8.25 for 2014 onwards, and we're hardly alone.

Here's an important reminder: More Americans (1.98 million) earn less than the minimum wage than earn the minimum wage (1.56 million), according to the Bureau of Labor Statistics. Tipped workers, like I was, earn a base pay of $2.13 an hour. They are expected to receive $5.12 an hour in tips to bring them up to $7.25.

Why I'm writing this post is to connect that importance of sub-minimum-wage workers to something else. Jodi Kantor had an excellent report last month on the role of erratic schedules for part-time workers who are also trying to care for a family. Kantor details the particularly painful experience of "clopening," that is, working until the close of a restaurant late into the night and then waking up early to reopen it the next morning, an experience I enjoyed a couple times.

Claudia Goldin's work has emphasized the role of hours -- in fact, she established a clear the flexibility of hours in an industry to the magnitude of the gender pay gap in that industry. It's likely that the tipped minimum wage is linked to these crazy hours. Let me explain why.

Let's imagine that I worked Saturday night, which is particularly busy on the Shore. Counting tips plus $2.13-an-hour base pay, I usually made well over the $7.25 minimum wage. But I'd often get assigned to work Monday mornings as well. Not many people like to eat seafood at 9 a.m. The consequence of this is that tips ran well below the $5.12-an-hour they'd need to be to keep me at a wage of $7.25 for every marginal hour. Instead, what was happening was that my tips from the Saturday night were being pulled forward in time, so that all I was really earning, on the margin, was the $2.13 an hour.

Some labor economists think that fraud and underpayment is one reason why the restaurant industry defends the tipped minimum wage so dearly. That's possible, but the story I just gave you provides the intuition for another reason why the tipped minimum gets defended: It gives the restaurant industry another pool of workers whose marginal wage is $2.13, because they've happened to pick up a shift that was more lucrative.

The key point is that this allows restaurants to stay open during hours when, at $7.25 an hour, it would not be otherwise profitable to be open. On the margin, those hours don't make sense -- they only make sense because the tipped minimum wage framework allows employers to average the cost of labor over a pay period. Blame the crazy hours, and perhaps the "clopener," on the tipped minimum wage.