July 1, 2003

This study aims to assess which of the top schools have the most "national" placement, as measured by hiring by elite law firms around the country.  This study proceeds on the assumption that "national" law schools (1) place large numbers of graduates at the best firms, and (2) place graduates at the best firms throughout the nation.

We studied the usual suspects for the top law schools-Yale, Harvard, Stanford, Chicago, Columbia, NYU, Michigan, Virginia, Texas, Penn, Cornell, Georgetown, Northwestern, Duke--plus two schools on the cusp of this elite group, Vanderbilt and UCLA. [1]   As a check on the reliability of the results, we added five very reputable, but presumably less national schools:  Emory University, Washington & Lee University, University of Notre Dame, University of Minnesota, and George Washington University.

To identify "elite" law firms, we used the Vault Guide to the Top 100 Law Firms, including the 23 "best of the rest" identified by Vault, top firms that didn't make the top 100.  In order to assess national placement power, we had to have a genuinely national sample.  Therefore, we studied only the top 3 firms in each city/region-- where there were at least three on the Vault list.  (The primary failing of the well-known American Lawyer study of hiring by the AMLAW 100 firms was that the sample was not national, with nearly one-third of the firms in New York City and more than two-thirds of the firms on the list in the Northeast corridor.  AMLAW 100 is informative as to job placement in New York and the Northeast, but says nothing about national placement power.)

Based on these selection criteria, we ended up with 45 elite law firms from around the country to study.  Organized by city/region, they are:

The Martindale-Hubbell on-line database was then utilized to determine the number of graduates of the 22 schools in the study at these firms.   School names were searched in common variations (e.g., "Univ. of Chicago" and "University of Chicago"), and also in forms to avoid false positives (e.g., "University of Texas," instead of "Texas," to avoid picking up, e.g., "Texas Tech").  Where the results showed zero graduates at a firm, the results were double-checked using additional search formats.

Because school size varies considerably, we calculated the "per capita" score for each school, where the "per capita" score represents the total number of attorneys divided by the total number of yearly graduates as determined by the Official ABA Guide to Approved Schools and information (e.g., about graduating classes, size of JD and LLM programs) on school web sites.  Because the Martindale-Hubbell search engine does not distinguish between JD and LLM graduates, it was necessary to factor in LLM graduates for purposes of the per capita measure, especially since some schools (Harvard, Georgetown, NYU, George Washington, in particular) have large numbers of LLM graduates (which may figure in the wide dispersion of Georgetown, George Washington, and NYU graduates in particular:  students come from all over for these LLM programs (especially in tax at Georgetown and NYU, and intellectual property at GW) and then return to where they come from).  (LLM graduates were added for all schools, however, in calculating the per capita measure.)

Keep in mind that because the search looked for all graduates at a firm (not just new associates), law schools noticeably more prominent now than a generation ago may not perform as one would expect (NYU is the most obvious example).

In addition to calculating the per capita score, we also calculated the number of firms in the study with at least one graduate of a school; and the number of firms with five or more graduates.  All these results appear below.  (Note that other schools not studied might well have outperformed Washington & Lee, Notre Dame, George Washington, Minnesota and UCLA in these categories.)  Finally, we prepared a chart showing the distribution of graduates at elite firms in each city/region, noting, where possible, the percentage of graduates placed at elite firms peculiar to that school's region (e.g., Texas/Southwest, California, Northeast corridor etc.).

The National Law Schools
(in terms of Elite Firm job placement)

40% of this rank is based on the per capita rate of placement in elite law firms around the nation for each school.  Another 40% of the rank is based on the number of elite law firms around the nation that have at least one graduate of the school in the firm.  Finally, 20% of the rank is based on the number of elite law firms around the nation that have at least five graduates of the school in the firm; this factor was given less weight since it slightly favors larger schools.  (The rank in each category is given in the charts that follow.)  All scores were normalized, with the highest scorer in each category receiving 100, all others a percentage of the highest score.  So, e.g., one can say, very roughly, that based on this data, Vanderbilt is "half as national" in its placement power and prestige as Harvard, and so on.

Other pertinent measures of school "prestige" and "quality" are included for easy comparison.

Five Things to Remember in Interpreting the Results

1. Self-selection by students in choosing where to go to law school in the first place has a lot to do with the results.  Students who choose to go to Stanford or Columbia are probably, among other things, choosing based on lifestyle and regional loyalties:  this is reflected, unsurprisingly, in where graduates end up practicing.  (How many students choose to go to Stanford in order to practice in Cleveland or Pittsburgh?  How many pick Columbia with the goal of big firm practice in Dallas or Portland?)  By contrast, Michigan, with only a minority of Michigan residents in its student body, doesn't offer many attractive regional options (no Detroit firms make the Vault lists, for example), with the result that its students both come from all over and spread out all over.  The same might be said for Duke and Vanderbilt (no North Carolina or Tennessee firms make the Vault lists, either).  Something similar is also true for Virginia, which combines both a low number of in-state students with proximity, and traditional feeder relationships, to attractive legal markets in D.C. and New York, among many others. 

   Contrast these cases with Berkeley or Texas, which have very high in-state student populations, are in states with major legal markets, and, like Columbia and Stanford, tend to be chosen by many students who have distinct lifestyle preferences and regional loyalties. 

   So the bottom line is this:  a student utilizing this data shouldn't conclude that they should choose Vanderbilt over Boalt-remember Boalt's faculty is better, its placement in law teaching is better (an even more elite selection process), and its students largely choose to be in California and environs, which has a lot to do with why they choose Boalt in the first place.  The same could, of course, be said for Columbia, NYU, Stanford, and Texas, among others.

2. Reputations die hard and are long in being born-especially among attorneys.  In 1970, the top five law schools were Harvard, Yale, and Michigan, with Columbia, Stanford, and Chicago fighting it out for the remaining two spots.  Penn was just on the cusp of the "top five," Virginia was clearly top ten, and then some mix of Duke, Northwestern, Texas, and Berkeley fought it out for the remaining top ten spots.  Cornell was surely top 15, NYU might have been top 15, Vanderbilt was surely top 20, and Georgetown might have been top 20.  UCLA was a brand new law school, just a half-dozen years old.

   Because this study looked at all attorneys at elite firms who graduated from any of the schools studied, it invariably reflects a school's reputation among elite firms over a long period of time, reflecting not only who Ropes & Gray was hiring in 1995, but some of whom they were hiring in 1970.  This goes a long distance to explaining NYU's showing, as well as Michigan's and Duke's.  (Duke also benefits from a highly inflated rating-relative to faculty quality-in U.S. News for most of the last decade.) 

3. Size matter-- at the margins.  Without a doubt, two of the measures used in calculating the overall rank are sensitive to the number of graduates (both JDs and LLMs), and in that category, Georgetown leads even Harvard and NYU (the next two largest) by a margin of more than 200 graduates per year!  If Georgetown were the size, say, of Penn, it is likely that it would rank more like Cornell in this kind of study.  Notice, for example, that Georgetown is distinctly underrepresented at the most elite firms in its hometown of Washington, D.C.

4. All 16 of these schools are genuinely national in their reputation and placement power.  The contrast with Washington & Lee (which, itself, is more national than most law schools) should make that clear.  (GW is on the cusp here, but its numbers are inflated by a large number of LLM graduates, and so does not deserve to be ranked with Berkeley and Cornell.)  And while big differences might be worth attending to (Michigan is twice as national as UCLA, but Duke is only slightly more national in its placement than Northwestern), the bottom line is that degrees from these 16 schools make you employable at the best private firms around the nation-and so you're better off choosing among these schools based on other considerations, like intellectual interests, special programs, cost, and lifestyle.  

5. Elite firm placement is only one measure of national reputation and standing.  Placement in law teaching jobs, an even more selective process, is revealing.  So is placement in clerkships, placement in the most sought-after public sector jobs, and so on.  Rank for law teaching placement is included above, and some data on clerkships is available elsewhere on this site (see Supreme Court Clerkship Placement), but systematic data on all these possible measures of national standing is not presently available.

Rank Based on Per Capita Score for Elite Firm Placement

Rank Based on Number of Elite Firms with 5 or More Graduates

Rank Based on Number of Elite Firms with at least 1 Graduate

 Elite Firm Placement by City

Remember:  the numbers below reflect only placement at (at most) the top three "elite" firms in each city, and so do not necessarily indicate what proportions of graduates go to those cities.  So, for example, Cornell, George Washington and Georgetown (see below) turn out to be underrepresented at the most elite firms in their main markets (New York and DC).  Note also that the figures for NYU, George Washington and Georgetown include large numbers of LLM graduates.

Keep in mind that the number of graduates in a particular market should be considered relative to the number of graduates trying to go to that market, not the total number of attorneys at these 45 elite firms.  Schools like Boalt and Texas-and to a lesser extent, Stanford, Columbia, and NYU, among others--have relatively small percentages of their class looking to leave the region, for the reasons noted earlier.

No regional placement percentages could be calculated for the following schools:  Michigan, Vanderbilt, Notre Dame, Duke.  In each case, no firms in the most important regional markets (Detroit, Nashville, Indianapolis, Raleigh, Charlotte) made the Vault list.

HIRING BY SOME VERY SELECTIVE FIRMS

Is Cravath, Swaine & Moore (New York City) the most selective firm in the country?  Perhaps, perhaps not.  But it is clearly among the most selective, and so its hiring practices shed some light on how it views law schools, especially those outside its region.  Cravath, unlike most firms, also makes it possible to search attorneys at the firm based on where they went to law school.  (Cravath also severely underreports attorneys-associates--to Martindale-Hubbell, which is why the figures here differ.  Of course, there is no reason to think the underreporting to Martindale-Hubbell systematically favors one school over any others, which is why that data is still useful for an overall picture.)  Here is where Cravath attorneys got their JD or LLM (there are lots of LLMs among the NYU and Georgetown numbers, for example):

Of selective firms in Washington, D.C., Kellogg, Huber, Hansen, Todd & Evans, a boutique telecommunications firm, may lead the pack:  their roster is packed with former Supreme Court clerks, law review editors, and attorneys with high-level government experience.  Kellogg, Huber is also small enough to make it possible, even without a search engine, to determine where their attorneys went to law school.  The smallness also, of course, affects the import of the results.

A larger, but still highly selective Washington, D.C. firm that has a search engine similar to Cravath's is Wilmer, Cutler & Pickering.  Here are all law schools with at least 3 attorneys in the D.C. office:

On the opposite coast, the elite Los Angeles firm of O'Melveny & Myers does have a search engine similar to Cravath which makes it possible to determine where O'Melveny goes to hire legal talent.  All schools with at least 3 attorneys in the LA office are listed.

A smaller Los Angeles firm, but, like Kellogg Huber, highly selective is Munger, Tolles & Olson.  Here are all law schools with at least two attorneys at the Los Angeles office of this prestigious firm:

When we go north in California, we find that the elite San Francisco firm of Morrison & Foerster also has a useful search engine.  What follows is a list of all schools with at least 3 (JD) graduates working as attorneys in the San Francisco office.  It is particularly striking here that regional schools like San Francisco, Santa Clara, and Golden Gate did not have as many attorneys here as non-regional schools like Minnesota, Emory, and Duke.

Unfortunately, none of the elite Chicago firms had appropriate search engines, and so no data could be collected for them.  We held off collecting data on elite Houston firms, like Vinson & Elkins and Fulbright & Jaworski, since the results are, to some extent, predictable.  Other firms may be added in the near future.