US News and World Report 2017 Economics Rankings

[Erin]

Indeed. Imagine what the reaction would have been if I had said something that was not carefully qualified! :-/

Exactly! With those qualifiers, it'd be hard to disagree.

 

I understand the sentiment, but IMHO it was a bit unwarranted.

[specious]

I don't understand why tm_member's claims are so controversial. The decision to attend a PhD highly depends on one's preferences and opportunity costs / outside options. Program rank is simply a heuristic for the quality of placements at those programs. Of course, prospective matriculants should carefully study placements, acknowledging the entire distribution of outcomes, accounting for attrition rates as well, and then make a decision based on the outcome of this analysis.

 

Fact of the matter is that few people become productive researchers in the end, even those graduating from the most prestigious programs. Coupled with the stress of a PhD program, the opportunity costs of losing out on work experience, and pursuing other graduate school options, such as business school, applicants should rigorously ask themselves why they should enroll in any program, and not why they shouldn't.

[chateauheart]

I mean, that user created a user-name called "upwardly mobile" (two days before tm_member made that comment), it's not hard to see why he was pissed about those statements.

 

Across the whole undergrad demographic in the U.S. and internationally, there are a lot of very talented students who have aspirations of moving upward in ranking during the PhD application process and further upward in the academic job market, and they tend to attribute all evidence/advice to the contrary as examples of elitism. It's hard to accept that for the vast majority of PhD applicants, your undergrad institution is usually the most prestigious institution you'll ever be in.

 

These sentiments from younger posters are understandable, and there's nothing much we can do aside from developing a thick skin and continuing to give sincere advice like tm_member has been doing in this thread.

[behavingmyself]

My claim was that students who are admitted to these kinds of programs should only attend if they have a good reason to do so, have no better options, have a fairly low opportunity cost, and know the likely outcomes. Looking only at those who do ultimately choose to attend those programs is not a valid test of my claim.

In what sense would you say that's different from people considering attending a program inside the top 50?

 

I would agree with statements like that the quality of training becomes noticeably worse shortly outside the top 50, and that the quality of the inside option falls more quickly than the quality of the outside option for the average student who is capable of gaining admission to a PhD program. But the thing of needing very specific reasons to attend a PhD applies to everyone.

[tm_member]

In what sense would you say that's different from people considering attending a program inside the top 50?

 

I would agree with statements like that the quality of training becomes noticeably worse shortly outside the top 50, and that the quality of the inside option falls more quickly than the quality of the outside option for the average student who is capable of gaining admission to a PhD program. But the thing of needing very specific reasons to attend a PhD applies to everyone.

 

Just inside the top 50, not so different. Top 10, different.

 

The only reason PhD programs exist is to prepare students for a career in research. They might not always succeed. Outside the top 50-ish, they succeed only very rarely.

 

Given that, unless you specifically want to work with one professor on a niche topic and you understand the risks, it becomes very hard to recommend them. The same is definitely not true for students admitted to Top 10 programs. As you say, as you move down the rankings the advice to be very cautious and as informed as possible becomes relevant.

 

Overall message: there are three tiers of PhD programs. Rankings within those tiers are less relevant than the differences between the tiers. The cut-off points are arguable when comparing two schools. When it is arguable, pretend like they are in the same tier for your decision. Below tier 3 is a risk only a very informed student should take.

 

Broader point that we all understand? The world is better characterized by continuity rather than discontinuity.

[kipfilet]

My 2 cents -- looking at a weighted average of placements in the last decade, the hierarchy would be something like

MIT ~ Harvard ~ Stanford > Princeton ~ Berkeley > Chicago ~ Yale ~ Northwestern ~ NYU > Columbia ~ UPenn

These 11 schools form a more consensual top10 these days. Once you move to the top20 there's much more room for discussion, as trends are much more noticeable (i.e. Minnesota and UCLA in decline, Duke and UCSD rising, etc.)

 

Also, I generally agree with the controversial statement: unless you are doing it for love/fun, it's not really a great investment of your time (and life) to go for a PhD outside of the top50. Bear in mind that your chances at an academic placement will be extremely low, and even the 3rd quartile-student will end up in industry. If you are risk neutral, the median placement is a good ex-ante indicator of where you will end ; if you are risk averse, look at the below median placement. If you look at placements for places that are almost in the top50, you can easily confirm what I said. Even policy institutions tend to only hire graduates from top20 programs (think IMF, Fed Board and the good regional Feds).

 

thus the original advice applies perfectly -- unless you have very specific personal reasons to go for it, you are better off heading to industry straight away.

[economicstyro]

Can someone please confirm what kipfilet says? "Bear in mind that your chances at an academic placement will be extremely low, and even the 3rd quartile-student will end up in industry?"

 

I personally have gotten a few offers from Top-70 and US News - unranked schools. These days I spent some time looking at the past placement results (if available) of my potential departments and it seems to me that at least 75% of their placements are academic, and that the overall results compare favorably with those from Top-50 schools. (Colorado-Boulder/Pittsburgh/Vanderbilt (US News top 50) as opposed to UT Knoxville, Binghamton, UConn, Washington State (US News top 70-80)). In my own observation, these lower-ranked (or even unranked) departments typically almost never have academic placements (excluding post-docs) into top-ranked departments, but their median students do get placed as well as those from higher-ranked programs.

 

In this vein, I would like to ask a relevant question: when it comes to placement, does the department's reputation or the advisor's reputation matter more? Thanks for your input!

[pocketecon]

Can anyone tell me where do the top UK and European programs feature in the top 50, if we were to categorize them using buckets or in absolute rank terms? There is lot of variability in the different rankings out there on the web for these universities. Specifically, can anyone tell me where would LSE and Oxford feature?

[Rohanps]

LSE probably in or around the top 10; UCL and Oxford around 30-40; Cambridge and Warwick maybe 50-60 at a push. All others below that.

[jklimek]

My 2 cents -- looking at a weighted average of placements in the last decade, the hierarchy would be something like

MIT ~ Harvard ~ Stanford > Princeton ~ Berkeley > Chicago ~ Yale ~ Northwestern ~ NYU > Columbia ~ UPenn

These 11 schools form a more consensual top10 these days. Once you move to the top20 there's much more room for discussion, as trends are much more noticeable (i.e. Minnesota and UCLA in decline, Duke and UCSD rising, etc.)

 

Also, I generally agree with the controversial statement: unless you are doing it for love/fun, it's not really a great investment of your time (and life) to go for a PhD outside of the top50. Bear in mind that your chances at an academic placement will be extremely low, and even the 3rd quartile-student will end up in industry. If you are risk neutral, the median placement is a good ex-ante indicator of where you will end ; if you are risk averse, look at the below median placement. If you look at placements for places that are almost in the top50, you can easily confirm what I said. Even policy institutions tend to only hire graduates from top20 programs (think IMF, Fed Board and the good regional Feds).

 

thus the original advice applies perfectly -- unless you have very specific personal reasons to go for it, you are better off heading to industry straight away.

 

I'm interested as to where this opinion comes from, that programs

 

I've looked at all the placement pages (when available) for programs and these are the conclusions I came to:

 

1. If I am the best student that has ever attended this school then I will place at a similarly ranked research university.

2. If I am a good student, I will place at a worse ranked research university (80+), a good LAC, decent government jobs, or a good industry job that pays very well.

3. If I am an average student, I will place at an average LAC, lower down government jobs (possibly state government), state universities, non-tenure track(VAP/teaching/lecturer) at research universities, or a decent industry job with more restricted options.

4. If I am a bad student, I will place at a LAC/state university in a bad location, or industry job.

5. If I am a really bad student, I won't be on the list.

 

When I say bad, I mean just a worse ranked student in the cohort. I'd be interested to see what more seasoned people on this forum think about these conclusions. I've also noticed that certain specializations tend to fall into certain types of jobs more, but I can't speak to that at all.

 

Edit: I don't know where post-docs fall on this list.

Edited March 22, 2017 by jklimek

[ksuga]

Hi Rohanps, how do you determine these comparisons? Oxford between 30-40? That would suggest schools like let's say UIUC, UNC, Ohio State, UVA which are ranked around top 30 in usnews are perhaps a bit higher ranked than Oxford?

LSE probably in or around the top 10; UCL and Oxford around 30-40; Cambridge and Warwick maybe 50-60 at a push. All others below that.

[tm_member]

Can someone please confirm what kipfilet says? "Bear in mind that your chances at an academic placement will be extremely low, and even the 3rd quartile-student will end up in industry?"

 

I personally have gotten a few offers from Top-70 and US News - unranked schools. These days I spent some time looking at the past placement results (if available) of my potential departments and it seems to me that at least 75% of their placements are academic, and that the overall results compare favorably with those from Top-50 schools. (Colorado-Boulder/Pittsburgh/Vanderbilt (US News top 50) as opposed to UT Knoxville, Binghamton, UConn, Washington State (US News top 70-80)). In my own observation, these lower-ranked (or even unranked) departments typically almost never have academic placements (excluding post-docs) into top-ranked departments, but their median students do get placed as well as those from higher-ranked programs.

 

In this vein, I would like to ask a relevant question: when it comes to placement, does the department's reputation or the advisor's reputation matter more? Thanks for your input!

 

I don't think you can view something like UConn's placement record (Ph.D. Placements | Department of Economics) as comparable to Pitt, CU Boulder, or Vandy.

 

I'm not going to spend too much time comparing these but for Vandy (https://as.vanderbilt.edu/econ/graduate/recent-placements.php) they have several US tenure track economics placements in good liberal arts colleges and research schools (2016 - Oberlin, Temple, UVA, U of Arkansas - Main Campus; 2015 - SUNY Oswego, Providence College, Miami). For Pitt the story is similar (2016 - Kenyon, Louisville, UNLV, 2015 - William and Mary, Colgate). Both Pitt and Vandy have several excellent international, industry, and government placements, too.

 

For UConn (Ph.D. Placements | Department of Economics) the placement record appears to be purposely misleading. Taking a brief look at the 2016 graduates, they are either not in econ (P. Yu), not research tenure track (M. Histen - clinical/teaching tenure track), or the listed university is not 100% clear. For example, H. Xu is working at Indiana University but at the satellite campus in South Bend, rather than the main campus in Bloomington. There is the world of difference between those. Even for SUNY Binghamton and UT Knoxville, the placement record is much better than UConn (https://www.binghamton.edu/economics/placement.html and http://econ.bus.utk.edu/documents/InitialPhDPlacement.pdf). However, the difference between the best outcomes at those schools and the best outcomes at Vandy and Pitt is still quite significant.

 

Don't get me wrong, there are good outcomes from those lower ranked schools. My claim is that those good outcomes should only be viewed as possible by someone who has carefully considered their options and has some knowledge of their own ability that was difficult to communicate or demonstrate to the adcom. Putting it one more way, if the best student form Harvard this year had instead attended SUNY Binghamton, they still would have done very well.

[kipfilet]

I'm interested as to where this opinion comes from, that programs

 

I've looked at all the placement pages (when available) for programs and these are the conclusions I came to:

 

1. If I am the best student that has ever attended this school then I will place at a similarly ranked research university.

2. If I am a good student, I will place at a worse ranked research university (80+), a good LAC, decent government jobs, or a good industry job that pays very well.

3. If I am an average student, I will place at an average LAC, lower down government jobs (possibly state government), state universities, non-tenure track(VAP/teaching/lecturer) at research universities, or a decent industry job with more restricted options.

4. If I am a bad student, I will place at a LAC/state university in a bad location, or industry job.

5. If I am a really bad student, I won't be on the list.

 

When I say bad, I mean just a worse ranked student in the cohort. I'd be interested to see what more seasoned people on this forum think about these conclusions. I've also noticed that certain specializations tend to fall into certain types of jobs more, but I can't speak to that at all.

 

Edit: I don't know where post-docs fall on this list.

 

Some things to bear in mind:

(i) Look at assistant professors at good LACs and where they got their PhD's from. I doubt you'll find many people with PhD's outside the top50. I went to the US News LAC ranking and randomly picked two top30 LAC's, to see where their recent hires are form (older professors do not count as the market was very different back then). Macalaster College only seems to have one AP, with a PhD from Minnesota. Oberlin's APs have PhDs from Columbia, Vanderbilt and UPitt. Even if you go down the rankings, you still find plenty of recent hires who have graduated from top20 schools.

 

(ii) Good industry jobs are more pedigree-obsessed than academia. This is especially true in finance (hedge funds, global banks, etc.). Goldman Sachs and AQR, for example, do not hire outside of the top10. It is true that there are some non-elite industry jobs that still require a PhD (i.e. transfer pricing), but many of the industry placements could be attained straight from undergrad (entering at a lower level, of course). Examples are major strategic consulting such as McKinsey and BCG.

 

(iii) Demand for policy jobs is likely to decline in the next few years due to the current political climate. The current administration has already stated that they tend to defund many federal agencies that are active in the PhD job market, heavily affecting those that do macro, labor, health, IO, etc. Also, the Fed System will cut substantially on hiring as they return to pre-crisis mode (the Board has publicly stated it had reached steady state after two years of hiring around 40 people a year). Furthermore, Dodd-Frank reform is likely to eliminate some agencies/responsibilities from other agencies and the Fed, further reducing the demand for PhD economists. All of this to say that as the policy market becomes thinner, it is likely that it becomes exclusively accessible to people from top departments (this is already more or less the case with the Feds and the IMF: the Board and the good regional banks only hire people from the top20 basically).

 

(iv) many places only report their best placements on their website. Even some top places do this (Harvard is known for that), but this becomes more pervasive as you go down the rank. So the placements you have looked at might only be the upper quartile of the distribution.

One good sign that this is happening is when a school lists 8-10 job market candidates in a given year, but then only report 2-3 placements per year. I just to the most recent US News ranking and picked two borderline top50 places randomly. First exhibit: Southern Methodist University has 9 candidates on the market this year, but only report 3 placements for 2015, 1 for 2014, 5 for 2013, etc. The second place I had picked was the University of Florida, but they neither report their JMCs not their placements on the website (or at least I couldn't find a link after 10 seconds -- huge red flag for me).

 

(v) Your conclusions sound pretty accurate for top50 departments, but I would doubt that the median dept ranked lower than 50 has such a positive outlook (see point above).

Edited March 22, 2017 by kipfilet

[Kaysa]

I'm interested as to where this opinion comes from, that programs

 

I've looked at all the placement pages (when available) for programs and these are the conclusions I came to:

 

1. If I am the best student that has ever attended this school then I will place at a similarly ranked research university.

2. If I am a good student, I will place at a worse ranked research university (80+), a good LAC, decent government jobs, or a good industry job that pays very well.

3. If I am an average student, I will place at an average LAC, lower down government jobs (possibly state government), state universities, non-tenure track(VAP/teaching/lecturer) at research universities, or a decent industry job with more restricted options.

4. If I am a bad student, I will place at a LAC/state university in a bad location, or industry job.

5. If I am a really bad student, I won't be on the list.

 

When I say bad, I mean just a worse ranked student in the cohort. I'd be interested to see what more seasoned people on this forum think about these conclusions. I've also noticed that certain specializations tend to fall into certain types of jobs more, but I can't speak to that at all.

 

Edit: I don't know where post-docs fall on this list.

 

 

Chances are you are not going to get the training needed to do good research outside a top 50 institution. I am pretty sure this opinion stems from the jmps from students outside top 50 institutions and the conference talks from faculty outside top 50 institutions. JMPs from students outside the top 50 consistently tend to cover outdated topics using outdated methods, and, in some cases, these students seem to be unaware of more modern methods. These characteristics make these students seem unprepared and careless, two labels that no graduate student wants to be associated with. Another source might be conference talks. Conference talks from faculty outside the top 50, and outside the top 100, in particular, tend to be worse in quality, and this probably carries with it a negative stigma.

 

I want to emphasize that I am talking about averages. Some students outside the top 50 are excellent as well as are some faculty. However, as you read more and more jmps and attend more and more conferences you start to notice patterns and that you don't want to waste your time anymore. Student packages start to get passed over and ignored. Sure, maybe you will find that diamond rough. However, I never have, and even if I did, there is always someone from a higher ranked institution that is going to be marketed by their advisers better, networked better, and ingrained in the system better.

 

It is very tough to compete with top students in top institutions. How do you compete with someone who works with an adviser who is pioneering the field, has access to unique data, can get you involved in projects with his/her top notch colleagues, can get you several early pubs, etc etc?? The playing field is totally lopsided and unfair.

[jklimek]

These were both really helpful, I appreciate the thought out responses. I did notice that pattern with SMU.

[Rohanps]

Hi Rohanps, how do you determine these comparisons? Oxford between 30-40? That would suggest schools like let's say UIUC, UNC, Ohio State, UVA which are ranked around top 30 in usnews are perhaps a bit higher ranked than Oxford?

 

Mostly via a combination of my own experience with their PhD students, my knowledge of the course and the professors there, and looking at their placement record over the past few years.

[pingpong]

Do you think programs like UNC, UIUC, Ohio are better than Georgetown? as is suggested by the ranking

 

I am confused, I thought Georgetown was somewhere between top 15-25, and that UNC was a top 40-50.

[tm_member]

Do you think programs like UNC, UIUC, Ohio are better than Georgetown? as is suggested by the ranking

 

I am confused, I thought Georgetown was somewhere between top 15-25, and that UNC was a top 40-50.

 

My ranking of those four schools would be Ohio>UNC>UIUC>Georgetown. I don't know anyone who would substantially disagree.

[ksuga]

I disagree that georgetown is between top 15-25, or even close to that for that matter. Please see page 11 in a summary of rankings over time from MIT. No georgetown in top 25.

http://web.mit.edu/ir/rankings/USNews_Grad_Rankings_1994-2016.pdf

Do you think programs like UNC, UIUC, Ohio are better than Georgetown? as is suggested by the ranking

 

I am confused, I thought Georgetown was somewhere between top 15-25, and that UNC was a top 40-50.

[pingpong]

My ranking of those four schools would be Ohio>UNC>UIUC>Georgetown. I don't know anyone who would substantially disagree.

 

Thank you tm_member and ksuga. So besides the ranking, it is also clear that UNC strictly dominates Emory?

[ksuga]

Yeah absolutely. Would like to know others views on the following 6 schools which tied 29th in the rankings. which of these are generally considered top 30: OSU, MSU, UIUC, UNC, Davis, UVA?

 

Thank you tm_member and ksuga. So besides the ranking, it is also clear that UNC strictly dominates Emory?

[pingpong]

Of course, any additional information will be appreciated.

 

Yeah absolutely. Would like to know others views on the following 6 schools which tied 29th in the rankings. which of these are generally considered top 30: OSU, MSU, UIUC, UNC, Davis, UVA?

[pocketecon]

Mostly via a combination of my own experience with their PhD students, my knowledge of the course and the professors there, and looking at their placement record over the past few years.

 

Can you please elaborate on your experience with their PhD students and the knowledge of the course and professors?

[kipfilet]

Of course, any additional information will be appreciated.

OSU ~ UVA > Davis > rest

[tm_member]

OSU ~ UVA > Davis > rest

 

Agreed with perhaps OSU > UVA by the tiniest of margins.