FRBNY RA: Do all former RAs get PhDs?

It has been two months into my new job at the Federal Reserve Bank of New York (FRBNY). As a research analyst (RA), I primarily assist economists with creating paper figures and tables, running regressions, and cleaning data.¹ It is common knowledge that many RAs move on to pursue a PhD since, in most cases, the RA job is meant to prepare for the things one is expected to do in a PhD program. I pose the following questions:

• What is the share of RAs that go on to pursue PhDs?

• What are the changes in that share across time?

Thanks to a RA at the NY Fed, I have a dataset (cleaned) that can help me answer this question. This blog post will describe the dataset and some statistics on the further education of former RAs.

All opinions are my own and do not reflect those of the Federal Reserve Bank of New York.*

The Data

The original dataset is a hand-collected data-set that contains 300 former FRBNY RAs across 18 years (2000-2018). Each observation contains information on a former RA’s name, function, start and exit date, email address, and a list, in chronological order, of jobs or higher education pursued. For the time series analyses, I use exit years since that is the year the former RA moves on to his/her next job or grad school. I do a further cleaning of this initial dataset (described in the next paragraph) to make the analysis easier to run in Stata (or in any statistical software of your choosing).

By hand, I parse out the list of post-FRBNY RA jobs/school and add the following new variables: highest education, where I only list the highest form of higher education a former RA has recieved (i.e. PhD > Master if they have both), highest education institution, subject studied, PhD dummy variable, job 1-6 (job 1 is where one went after FRBNY RA job, job 2 is where one went after job 1, so forth and so on), and four indicators indicating the types of jobs a person has gotten. The categories for the job indicators are

(1) Academia - any Professor, RA (not including FRBNY), or research job (e.g. Brookings)

(2) Private Sector - any private firm or organization

(3) Government - either national or state government work

(4) Central Banking - because of the gray area, we list central banking in its own category.

It is important to note that based on the current dataset, only those up to 2012 would have completed their PhDs if I assume that a PhD takes on average 6 years to complete (2012 PhD student would hit the job market in 2018). Hence, for those that exited the FRBNY RA position after 2012, I assume that they are still in school unless it is recorded they have dropped out of the program.

The dataset does not contain information on prior employment or education. Historically, there have been few instances of FRBNY RAs coming from other private sector or research jobs, such as JP Morgan, other Federal Reserve Banks, or think-tanks, but a majority of FRBNY RAs take the job straight from his/her undergraduate education.²

For now, I will only look at variables that involve higher education and job 1-6 (that is, did the former RA move onto grad school immediately or another job before eventually going on to grad school).

Results: To Grad School or not to Grad School?

This section describes some summary results. Since this dataset is basically a survey dataset, there may be non-reporting biases. I am still trying to figure out ways to deal with the potential missing data, but for now, I will only use what is given to me based on the survey.

Share of PhDs vs Higher Ed

First, I look at the share of former RAs that pursue higher education. Higher education includes pursuing PhD, MA, MBA, MPA, MPP, MPH, Mphil, JD, or MD degrees. Furthermore, I let PhD be the higher education degree of interest.

Across 2000-2018, 75% of former RAs go on to pursue high education at some point beyond working as a FRBNY RA, with 52% of former RAs go on to pursue a PhD. Figure 1 depicts the time series share of former RAs that pursue higher education (with those that pursue PhDs), between exit years 2000 and 2018. Note that for this first exercise, I do not account for timing of when they receive their higher education. That is, in Figure 1, for example, getting a PhD right after being a FRBNY RA is the same as getting a PhD two years after leaving the FRBNY RA position.

Figure 1: Time series share of former RAs pursuing higher education (and PhD).

The share of former RAs pursuing some higher education remains pretty consistent across the years, with dips in 2004 and 2011. The share of former RAs pursuing PhDs generally increase over time, but takes a dip in 2004, 2007, and 2011. For most years, the share of former RAs that pursue PhD tracks that of former RAs that pursue higher education. I add recession shading in hopes of explaining some of the noise in the time series shares. During the start of the recession (exit year 2007), the share of former RAs pursuing PhDs (decreasing) move in the opposite direction compared to those pursuing higher education (increasing). This implies that former RAs, during the beginning of the recession, would rather pursue some other form of higher education instead of a PhD. An explanation for this trend could be a panic shock, where those that do not desire to pursue a PhD will pursue some higher education due to higher levels of unemployment in the private sector. These trends move in the same direction once again in exit year 2008. It is safe to say that getting a PhD, compared to getting some other form of higher education, after being a FRBNY RA becomes more appealing over time. This can be seen from the smaller distance between the two lines over time.

We can take a moving average of the shares from figure one to smooth out the year-to-year fluctuations. Figure 2 depicts the shares of higher education (and PhD) with a 3-year moving averages.

Figure 2: Time series share of former RAs pursuing higher education (and PhD) with 3 year moving average.

The moving average is calculated by taking the average shares of years t, t-1, and t-2. The increase in share difference between higher education and PhD during the recession years is much more obvious. Also, the rising popularity of a PhD shown by an increase in the shares in recent years is also more obvious.

Share of PhDs vs Higher Ed Straight Out of FRBNY RA Job

Next, I look at the timing of when former RAs go on to get higher education. Here, job1 is the variable of primary focus: do RAs pursue higher education (or PhD) right after the FRBNY RA job? I calculate the following metrics both aggregately and by year:

(1) Share of FRBNY RAs that pursue higher education directly after FRBNY RA job vs those that pursue higher education at any point in their career.

(2) Share of FRBNY RAs that pursue PhDs directly after FRBNY RA job vs those that pursue PhD at any point in their career

Across 2000-2018, 81% pursue the higher education right after being a FRBNY RA, with 88% of former RAs that pursue PhDs work on their PhDs right after being a FRBNY RA. Figure 3 depicts the time series share of higher education former RAs that pursue higher education right after leaving the FRBNY RA job (also the share of PhDs that pursue PhD work right after leaving the FRBNY RA job).

Figure 3: Time series share of higher education former RAs that pursue higher education right after being a FRBNY RA (and that for PhDs). Note that PhDs are still a subset of higher education here.

We see that more and more former RAs are pursuing higher education right after their FRBNY RA job. This is especially true for those that pursue JD. Working in the opposite direction, many former RAs go to another job before pursuing a MBA. Looking specifically at PhDs, we see that former RAs that pursue PhDs consistently choose to do so right after their FRBNY RA job.

Once again, I include recession shading to possibly give us insights into the trends we see. Looking at exit year 2007, we see that there is an increase in the share pursuing higher education right after the FRBNY RA job. This trend is also evident when looking at PhD shares. Here, we can more precisely pinpoint the shock mentioned earlier, and this almost similar increase implies that the trends we saw in Figure 1 are actually not as significant as they may seem. Furthmore, over this recession period, more former RAs that pursue a PhD do so right after the FRBNY RA job. However, it is unclear if this is due to the recession or not.

The time-series shares in Figure 3 for years closer to 2018 are misleading because we haven’t observed enough years for those that end up going to the private sector. For example, if a FRBNY RA leaves for the private sector in 2017, they might choose to pursue higher education (either PhD or not) some time beyond the dataset, which changes the share by including one more FRBNY RA in the denominator. As more data is collected for subsequent years, the shares for years closer to the end of the dataset might hover near the mean.

In conclusion, only a slight majority of former RAs go on to pursue a PhD. However, for former RAs that do pursue a PhD, the majority do so directly after the FRBNY job. A large share of former RAs pursue a higher degree, including PhDs. This result is meant to highlight this key insight: while most FRBNY former RAs go back to school, it does not necessarily have to be a PhD. There is variation from year to year, possibly due to outside events such as a recession, but the general trend highlights that PhDs are becoming more popular for FRBNY RAs across time while there is a consistent high share of FRBNY RAs pursuing higher education.

Do all FRBNY RAs become economists?

For former RAs pursuing PhDs, most pursue an economics PhD. Across 18 years, of all former RAs that pursue PhDs, 85% of them go on to pursue economics PhDs. These include Economics (via the Economics department - could be pure microeconomics or macroeconomics), Business Economics (via the Business School), Financial Economics, or other more niche Economics (i.e. Health, Agricultural, Political, etc.). With the remaining 15% of former RAs that do not study economics, they study other social sciences (like History or Sociology), business organization and marketing, STEM (like Physics and Computer Science), and mathematics/statistics. There are 2 observations where the subject of the PhD is unknown. Figure 4 graphs these shares in a pie chart.

Figure 4: Breakdown of PhD subject areas into 6 categories.

Placement of Econ PhD programs

Many FRBNY RAs get placed in top PhD programs. Here, I will only look at the top 10 economics PhD institutions using current US News rankings. I assume these rankings have not change too much during exit years 2000-2018. Furthermore, I aggregate the economics department with the business school (i.e. Harvard Business counts as Harvard Econ., MIT Sloan counts as MIT, etc.). Table 1 presents the number of former FRBNY RAs in each of these top institutions:

Table 1: Placement of FRBNY RA in Top 10 Economics PhD Programs

From the dataset, 75% of all FRBNY former RAs pursuing a PhD get into a top 10 Econ PhD Program. Those are very promising placement results for FRBNY RAs!³

Conclusion

Based on the preliminary statistics and figures, not all FRBNY former RAs go on to get PhDs. In fact, across 18 years, only 52% of the former RAs go on to get a PhD. Of those that get PhDs, 85% go on to pursue an economics PhD. 75% of the former RAs go on to get higher education of some sort. Looking at the timing of when a former RA pursues higher education, 81% of them pursue higher education directly after the FRBNY RA job. Of all former RA PhDs, 88% pursue their PhDs directly after the FRBNY RA job.

A large majority of FRBNY RAs pursue an economics PhDs (85% of those pursuing a PhD). In terms of economics PhD placements, 75% of them get into top 10 economics PhD programs.

There is still much work to be done with cleaning the dataset and finding ways to deal with missing survey data. There will be more analyses to come, especially looking at trends within function and the categories of jobs that former FRBNY RAs hold after their tenure at FRBNY. Let me know what else you all want to see.

^{* I would like to thank Sonya Chen and Brendan Moore for the helpful comments and feedback.}

^{1 I work in the Financial Intermediation (FI) function. The fuction primarily study financial intermediaries (i.e. the institutions that are between the lender and the borrower) and bank regulations.}

^{2 Just speaking from personal experience. I took the FRBNY RA job straight from my undergraduate studies.}

^{3 Of course, this is assuming one does good work, puts together a solid application, receives good recommendation letters, etc.}