How the "Cravath System" Created the Bi-Modal Distribution

The bi-modal distribution (graphic to the right) continues to generate interest in the blogosphere.  See, e.g., Greg Mankiw, Right Coast, Broken Symmetry.   The  chart summarizes the starting salaries for lawyers who graduated from law school in 2006.  One reason the bi-modal structure is so jarring is that it demonstrates that measures of central tendency, such as average or median, are not necessarily reliable guides for law students' future earning power.  In conventional labor markets, that disconnect is rare.

NALP recently dug into its archives to determine whether this stratification is a persistent feature of the entry level law market.  See NALP Bulletin (Jan. 2008).  It turns out that 15 years ago, the market followed a much more traditional distribution.   The chart below summarizes the salaries for the class of 1991.

The 1991 graph is right skewed but bears some resemblance to a normal curve.  Below is the graph for 1996:

The rightward skew is a bit more pronounced and the area under the $75K to $85K range is becoming more substantial.  A more seismic shift is seen in 2000 (below) with the emergence of a second mode at the $125K price point. 

At the height of the Internet boom, a remarkable 14% of all entry level lawyers took jobs at the $125K level.  According to NALP, "never before had a single salary so dominated the landscape."  Under the 2006 bimodal distribution (see chart above), 44% of graduates received entry-level salaries in the $40K to $60K range; yet, the second mode moved further to the right ($135K to $145K) and grew to 17% of all graduates.

What are the market forces that have created this peculiar salary structure?  In my working paper, "Are We Selling Results or Résumés?: The Underexplored Linkage Between Human Resource Systems and Firm-Specific Capital," I posit that the runaway $160K mode is a confluence of two factors:  (1) the continued growth in the corporate legal services market, primarily due to the growing scale and scope of transnational corporate activity; and (2) law firms' nearly universal adherence to the "Cravath system," which purports to hire the best graduates from the best law schools and provide them with the best training.  More after the jump. ...

The New York firm of Cravath Swaine & Moore created and refined this system during the early 20th century.  The emphasis on educational credentials was initially an attempt to establish a distinctive brand of legal services that could differentiate the firm from other Wall Street competitors. Now, ironically, it has become a uniform industry practice utilized by every large law firm that claims to provide first-rate services.   [Virtually all firms mimic the Cravath system without understanding its logic.  In the paper, I draw upon a unique study of engineers at the renown Bell Labs to suggest that Cravath's superior client service has less to do with credentials than an organizational structure and ethos that aligns the interests of associates, partners, and clients.  Paul Cravath's theory is laid out in Robert Swaine's history of the firm.  I will discuss his profound disconnect in a subsequent post.]

On one level, law firms' reluctance to tinker with the Cravath system makes perfect sense--it has produced large incomes and huge profits margins for decades.  Further, 30 or 40 years ago, the vast majority of  firms that would eventually become the Am Law 200 were, in fact, "white shoe" firms within an overwhelmingly regional corporate legal market.  In particular, places like Cleveland, Detroit, Pittsburgh and St. Louis garnered their share of elite law school graduates.  In the early 1960s, sociologist Jack Ladinsky found that 73% of Detroit lawyers working in law firms (i.e., not in solo practice) went to one of five national schools:  Harvard, Yale, Columbia, Chicago, or University of Michigan.  See Ladinsky, Careers of Lawyers, Law Practice, and Legal Institutions, 28 Am. Sociology Rev. 47, 49 (1963).  You can bet this pattern is radically different today.

As these regional law firms morphed into the Am Law 200, their partners remained psychologically wedded to their own perceptions of eliteness.  In the ensuing salary wars, these firms slavishly paid the prevailing rate rather than signaling to the market that the firm had become "second rate" (a term used by a Proskauer Rose partner in rationalizing the higher pay). In turn, the laws of supply and demand produced the bi-modal distribution.

The Results or Résumés paper draws upon two pieces of market data to demonstrate that a large proportion of large corporate law firms have to re-evaluate their business models:  (1) stunning uniformity of associate entry level salaries amidst large, growing disparities in profits per partner; and (2) evidence that firms are becoming stratified by premium versus non-premium practice areas.

Regarding the disconnect between associate and partner pay, the bar chart below compares associate starting salaries with profits per equity partner at the 25th, 50th, 75th, and 95th percentile breakpoints in the Am Law 200.

As the paper documents, over the last several years, profits are going up at all levels of the Am Law 200; they are just going up much faster for high PPP firms. Obviously, when firms at the top of the heap (95th percentile) pay higher salaries and bonuses, it is quite a stretch for firms at the 25th percentile to match.  The money has to come from somewhere.  If it comes out of the draw of a rainmaking partner, he or she has a strong incentive to seek greener pastures.  That is risky; but to the majority of the partnership, departing from the Cravath model seems equally perilous. Over the last decade, virtually all large firms have adapted by increasing leverage (i.e., the ratio of total lawyers to equity partners).  But higher leverage can also undercut associate morale and loyalty.  So for firms at the bottom of the PPP distribution, the salary wars are one hell of a vise, particularly as the economy heads south and they are stuck with a $160,000+ cost structure.

Yet, for many firms, there is second trend that is much more troubling.  Based on a dataset of lateral partner mobility within the Am Law 2000, it is possible to tease out a relative hierarchy of practice areas.  The table below, which covers the 2000 to 2005 time period, orders legal specialty by differential profits per equity partner (PPP) between the firm a partner left and the firm he or she joined.

The trends are straightforward. Partners in marquee practices like white collar crime, securities enforcement, M&A, private equity, emerging markets, and intellectual property litigation are disproportionately moving upstream to more profitable firms. Partners specializing in regulatory compliance, real estate, public finance, project finance, and trust & estates are disproportionately moving downstream.  A similar analysis using multivariate regression, which controlled for year and city, found that labor & employment was also associated with downward (i.e., lower PPP) movement. (See regression table.)

In the long-run, firms without a optimal mix of premium practice areas will have a hard time sticking with the Cravath system.  Increasingly, corporate clients are refusing to have their cases staffed by expensive first- or second-year associates who don't know very much and tend to leave.  Hence, the training the clients are allegedly paying for has little or no future payoff.   

In other words, for many large law firms, the wheels of their hallowed business model are falling off.  During this period of denial, every firm's short term strategy is to work harder, promote fewer lawyers to equity partner, and de-equitize as needed.  Marc Galanter and I chronicle the unremitting nature of modern large firm practice in our forthcoming Elastic Tournament article.  If you have any doubt about the inevitability of change, read this seminal speech by Cisco GC Mark Chandler.

Fortunately, the Results or Résumés paper lays out a solution for any law firm willing to try something new.  The psychological barriers, however, are much larger than the logistical or financial.  I will blog on this topic in a subsequent post.

Comments

[Archana]

This is very interesting, Bill, and certainly illustrates the salary distribution I recall hearing about when I graduated in 2001. I had no idea it was such a new phenomenon. I look forward to reading the working paper...

[recent ls grad]

Incidentally, the "Cravath System," at least as it's pitched to potential incoming associates these days, refers to Cravath's practice of assigning each associate to a (approximately) a pair of partners for about 18 months, and then switching him/her to another set in a different focus area (within the corporate or litigation dept.). Their system, so the story goes, is to be contrasted with other firms that pigeonhole associates as quickly as possible.

Indeed, your statistic that in the early 60s "73% of Detroit lawyers working in law firms . . . went to one of five national schools" makes me skeptical of the notion that Cravath really invented prestige-whoring at large law firms. Were these Detroit firms really looking to Cravath in making those hiring decisions.

Anyway, any explanation of the current nature of large law firms has to begin with the Securities Exchange Act of 1934 (and probably end there, too).

[Vickie Pynchon]

when a partner at a second tier firm in the year 2000 I repeatedly suggested we abandon our expensive summer associate program and stop hiring 1st and 2nd year lawyers , letting the amlaw 100 train them. It would be suicide I was assured. We were acquired after I left. Suicide by merger.

[Jeff Yates]

[cross posted on Voir Dire Blog]

1. Notwithstanding the tendency toward a more central salary distribution in 1991, I recall a curious lack of linearity to relative credentials and the market at that time. It seemed that while top 10% to 15% students had great opportunities, the correlation between class rank and opportunities tailed off considerably with the remaining students. In other words, there wasn’t a huge difference between being in the top third and the bottom third. This is certainly an unscientific observation and limited to one person’s experience.

2. It seems that large law firms would be better served by simply waiting around a couple of years to see how new lawyers developed and then cherry picking the best ones from firms that can’t compete with the salaries they could offer. Compare this with the ridiculous amounts of money, time, and resources devoted to summer associate programs that are arguably not even used for evaluation anymore since nearly everyone gets an offer. Perhaps this odd situation is driven by a difficulty in systematically measuring lawyering ability beyond grades or prestige of degree granting institution.

3. Bill seems to imply that the Craveth model (and its accompanying salary distribution) is not sustainable. I agree that it seems a bit nonsensical, but how is it not sustainable? A lot of inefficient, unfair, and silly systems persist for very long periods of time. What is the mechanism by which it will fail? Will clients run away from these firms to small or medium firms with specialized practices? Will the “high end” practices such as securities and white collar criminal defense splinter off into boutique firms and leave the remaining practice areas partners high and dry? What about lawyer supply systems - will undergrads quit choosing to go to law school?

4. Jim Chen has a nice follow up post on Moneylaw here, naming Bill’s post as the “post of the year.”

[Corey]

I think the associate to partner pay chart is neat, but like many things in the legal market, it tends to over-emphasize the dramatic results at the topmost tier.

In any other universe, a business that can make 3 or 4 times salary in revenue on a first or second year employee is doing wonders. The fact that Wachtell and Cravath can skim $2M per partner off the mergers and bankruptcies of the world's leading brands doesn't mean that all firms have to run on that model. The fact that Wachtell and Cravath can motivate law grads to bill 2500 hours on the same salary everyone else gets is just a function of elitism and gullibility, and has nothing to do with the question of whether paying a new lawyer $160K to bill 1800 hours at $300/hr ($540K) makes sense or is sustainable.

Here is the positive spin:

Billing rates have gone up with salary increases. The salary/revenue ratio for an associate working X hours is the same at $160K as it was at the $125K level. Of course the associate and the firm are splitting a bigger pot so both make more.
Maybe some clients grumble about paying new law school grads $300 an hour, but that second mode on the chart is pretty narrow. Going down to the first mode isn't an option for most GCs, who fear having to go to their boards with a lost big case/deal after choosing "second rate" attorneys.

And here is the negative view:

Of course the wheels are falling off. Of course you are right that the firms are asking for more work and making fewer partners. Attrition is up, layoffs are no longer unheard of, work is getting outsourced to contract attorney sweatshops. Morale is down, inspiration is down. People with ideals are marginalized for the sake of the bottom line. Everything and everyone is fungible. Loyalty (from both sides) is incommensurable thus suspect. Associates have been quietly and unawares converted from professionals to wage slaves.

Is anyone suprised? The same process has happened to engineers and to just about every other profession over the last 30 years. As far back as high school I have been called a gen X slacker while working away my life and watching the boomer generation gradually consolidate wealth in a smaller and smaller demographic. There is a grand pessimism that has been wrought on my peer group. The unaware or cynical ones step lively for Cravath and get picked up to help perpetuate the system. The rest just find a reasonable job to pay the bills and try not to comprehend the brutality of it all.

Is that sustainable? Yes. It runs on mythology. Hard work makes good. Everyone has a shot at a Cravath partnership if one is just smart, dedicated, and plucky enough. American Dream 101. If that guy can do it so can all hundred of us!

One commentor pointed out that summer programs are pointless if everyone gets an offer. For the amlaw 100 I disagree. How many filet dinners at Partner X's Hamptons estate does it take to make a 24-year-old dump his sweetheart and sign on for 2500 billable hours? The summer program is a pitch too.

[Corey]

Congrats on the Stanford placement for your Elastic Tournament article! One note on it: I took a career detour but I am in the "millenial" lawyer peer group (as my comments above will also prove.) Your article's description of "us" sounds like it was written by and for boomers. In reality, we end up working as hard or harder than the boomers did, just without the blinders on. (Nothing like watching your dad lose his 30-year job and pension in an "economic" layoff to give a little perspective.)

The walk to school in the snow gets longer every time the story gets told. I have a hard time buying that these partners worked so much harder for so much less when I start historical pricing houses and lifestyles several decades back.

I wouldn't attribute any difference between boomers and millenials to work ethic, but rather to the gradual failure of the Horatio Alger myth as a motivator, when read against an increasing postmodern skepticism and alienation. I know how to motivate my peers to work hard, but I have yet to find a partnership or a client quite that progressive.

Anyway, a side point on an article that will undoubtedly create much interest.

[Michael F. Martin]

If you integrate that bimodal distribution up into a cumulative distribution, then you get something that looks a bit like a Friedman-Savage double-inflection utility curve, no?

If I remember right, Friedman and Savage thought this kind of function suggested class differences.

I'm not sure I agree with their interpretation. The double-inflection utility curve could be a cross-section of a two-dimensional surface showing concavity and convexity over different periods of time.

[biglaw partner]

Luckily, this is just a blog comment, so that rule that "only those who have done the reading are entitled to participate in the discussion" doesn't apply. I'll try to read the paper later, but, in the meantime, here goes.

I would think that the most likely development for first-year associate salaries in the foreseeable future (10 to 20 years) would be a continued spreading of the top hump of the distribution, as second-tier firms make less of an effort to match the very top tier. To some extent, this has already happened: a chart that included first-year bonuses would reflect a considerable difference between first years at, say, Simpson Thacher and DLA Piper.

Other than that, I don't see how the charts or the analysis indicates that the "wheels are falling off." Any time there is a wave of salary increases, corporate GC's complain about first year associate pay at biglaw. There isn't any real evidence that I know of to believe that they are generally refusing to have matters staffed by junior lawyers. Also, the chart of ratios between first year pay and equity partner pay is interesting, but it doesn't prove that the current situation is unstable. It's common for the junior people in a field to have similar incomes while there is substantial divergence at the top. Take a look, for example, at the pay of baseball players or musicians. Or professors, I think.

[Michael F. Martin]

Mr. biglaw partner:

The "superstar" distribution of salaries among musicians and athletes could have the characteristics you describe without being bimodal. For example, they could be poisson distributions.

The bimodality is not definitive proof of instability; but it's not good!

[HAL]

1) I vehemently disagree that young associates are overpaid. You devote no treatment to the rising cost of tuition at the elite institutions preferred by the AMLaw 100. Total cost of a JD is about 200k at these places. How many years *should* it take to repay? How much did it cost in 1991?I think part of your analysis may be that it's overbroad to discuss the top 200 firms; I'd narrow it to the top 100 if not 50. At which point, you realize that billing rates for first-year associates are even higher, and if an associate bills a standard 2000 hours, 160k is actually pretty low given that firms stand to make 6-700k+ in revenue on that associate. Isn't that an unusually low labor share if you model production on a douglas-cobb model?

2) I believe it's "achilles heel" and not "heal", free edit, page 22 =)

3) Doesn't Wachtell do so well in part by charging success fees for M&A deals that are a % of the transaction, similar to financial advisors? I think that may fit with the model you propose. Similarly, I'd be more interested to see an exploration of areas in which cost pressures are intense -- labor and employment, for one. What might these alternative arrangements look like? Have some parties adopted them?

4) Ultimately, I feel like the inexorable economic conclusion is that there are too many lawyers and they waste too much time. Hypothetically, corporation ABC retains XYZ to handle all matters of labor and employment, for, say, a fixed price per year plus a fixed amount per dispute or executive hire. The only way for such an arrangement to be profitable is: XYZ has to spend less time (billable hours) solving these problems, hires fewer attorneys (or pays them less) and shares the savings with ABC (in the form of the fixed prices). Otherwise, it's just cost-cutting by another name. Either way, less money, fewer lawyers. Fine with me, I guess; it mostly just strengthens my conviction that the ABA should decertify every law school outside the top 50.

[Russ]

Why are legal fees and legal salaries increasing at all? With all the technological improvements, why aren't legal fees going down? Lexis/Westlaw, the internet, computers, word processing--all these technical innovations make it far easier and faster to do legal work than ever before.

Shouldn't the technological improvements be exercising downward pressure on the costs of legal work?

As a companion question, why are lawyers working longer hours than ever before when all these technological improvements make their jobs infinitely easier than it was in 1950 or 1905?

[Graham]

Russ,

I think that Westlaw and Lexis have created what some older partners are calling the "westlaw generation" - and I don't think that it is a compliment. The accusation is that young lawyers simply search for sound bites and don't read cases any more. And obviously they are right.

If you look at a world without computers and the internet, lawyers wrote less and reasoned more. Now, unfortunately, there is a case somewhere out there (my "bosses" are "sure") that will support nearly any position. And young associates think that if they just spend enough time in front of a computer screen, they are going to find it.

But this shouldn't surprise anyone - the Supreme Court is an obvious example of legal blunderbus gone crazy. It shouldn't possibly take 40 pages to write every opinion. And neither do we need 40 page dissents to go along. Lawyers for the most part are simply people that want to shoot verbal bullets at each other, and the internet has simply given us more ammo. Its a shame.

[Corey]

"With all the technological improvements, why aren't legal fees going down?"

Because clients are no more paying hundreds an hour for lawyers to look stuff up on the internet than they were paying a few less hundreds "back in the day" for lawyers to look stuff up in books. Clients are paying for brand names, attorney client privilege, malpractice coverage, and deniablity.

Is that cynical? yes. True? yes. Imagine you want to force a competitor out of business. You can pay firm X five million dollars for years of highly technical patent lawsuits, or you can pay Cravath one million for an opinion letter, a threat letter that says "Cravath" on it, and a settlement offer phone call from a partner who is on the board of two fortune 500 companies. Top firms are not living or dying on the quality of any service capable of being improved via technology. Prestige is what is being sold, and that explains the bi-modality in my view.
It also explains why amlaw200 new associate salaries are not differentiated by talent/merit/prestige between the firms but profits per partner are.

[Mike McDougal]

This talk of books and Lexis/Westlaw is horribly misguided. The costs of legal research are a miniscule portion of the cost of providing legal services. So, yes, Russ, the cost of legal research is a downward pressure. But downward pressure on a miniscule input doesn't change the total cost much. The vast majority of the cost of legal services is labor.

[Michael F. Martin]

Mike McDougal is absolutely right. It's the labor that's the key. And it's the QUALITY of the labor that is the sole source of competitive advantage for law firms. Without a perception of quality, there is no way for law firms to pass increasing costs on to clients in a competitive market.

So partners are forced to bid higher for the top law grads in order to attract more of them to do more work -- and thanks to globalization there's never been more work than now.

...and then associate to partner ratios have to go up to keep the profits per partner high enough to prevent partner defections.

Which is fine for a while... until associates start to come up for partnership who have never met clients because they've spent the last decade doing document review or due diligence. Whoops.

Kudos to Bill Henderson for pinpointing lack of training as the main problem for stability of the current business models. Big firms can dress it up as much as they want; but classroom lectures do not make up for hands on experience with clients or in court.

[Associate]

Westlaw/Lexis have probably increased costs because of the completely unreasonable expectation of partners in the pre-Westlaw generation who believe that absolutely anything can be found with a simple search. I've spent hundreds of hours searching for things that simply do not exist because a partner wants to include a completely unsubstantiated claim in a brief.

[JRH]

Lots of good points, but I'd like to echo a few contrarian points above:

1. Where are the GC's going to go? These people owe fiduciary duties. If things turn out badly, a GC is going to want to be able to say that he hired a reputable -- albeit expensive -- firm. Why would a GC ever go for the "low cost leader" and put himself on the line?

2. Your post, while excellent, completely overlooks the cost of law school. Twenty years ago salaries looked far different; let me assure you, law school tuition was different as well. Most students graduate with six figures of debt.

3. First-mover bias: firms that cut associates (e.g., Sherman) or pay below market rates will lose face. Period. Law is a prestige oriented field and reputations change slowly. When profits are booming, why fight the status quo?

An even more interesting and relevant article would focus on the dubious tactics lower-ranked schools use to perpetrate the (mistaken) belief that their students will emerge on the good side of your bi-modal "curve".

[Spartee]

People who cite law school tuition as a factor are mixing up the causation relationship. Law firms do not price the salary paid to new hires based on the college once tuition paid by the new hire. Rather, the law schools are more likely basing their tuition with an eye toward what new graduates can make. (As a U of M grad, I swear by all that is holy or otherwise never shall that school get a kind word from me or a donation.)

JRH is also incorrect, I think, when he says law firms lose face when cutting associates. No one cares or remembers one year later. When people decide whether to take an offer, the bad outcomes for your predecessors are discounted, explained away, denied, or even weighed against the opportunity presented. Look, if the Marines can get 18 year olds to keep signing up, even though many graves of prior offer-takers are conspicuously displayed, how hard is it to find 25 year olds willing to take $160k?

People are spot on about closing all those law schools below the top 50-100, though. Those places are a disgrace.

[ASlyJD]

Yes, how absolutely horrible that small markets have a local law school to turn out grads to the local market and to centralize CLE instruction.

I can see the argument for closing law schools in cities that have four or five. (Boston, New York, D.C., Chicago) But here in flyover country, if the schools ranked over 100 get closed, goodbye to having more than one law school in a state.

Going to one of these schools, I realize that without an LLM from a "prestigious" school or years of high class experience I will never see the inside of big prestigious white shoe firm. But that doesn't mean that my education is worthless or that my school should be closed simply to create an artificial shortfall.

And really, has US News and World Report so brainwashed people that a school's reputation is only as good as a number?

[Andy Freeman]

> Look, if the Marines can get 18 year olds to keep signing up, even though many graves of prior offer-takers are conspicuously displayed, how hard is it to find 25 year olds willing to take $160k?

Part of the allure of the Marines is that there are conspicuously displayed graves. The Marines offer meaning and significance.

[Think38]

Interest article.

When looking at whether the ABA should decertify law schools, please keep in mind that most lawyers don't work for the Amlaw 200 firms or practice corporate law. Large numbers help rather ordinary people with very real life problems like lease disputes and traffic violations. There's nothing wrong with that.

Second, the impact of technology is a double edges sword. Sure, computerized research cuts down on research. But modern word processing now makes it possible to easily make a partnership agreement take up 75 pages that used to be accomplished in 15 pages. Provisions that were never addressed a generation or two now contain terms with great specificity. The result is creating or reviewing one of these agreements (or a purchase or merger agreement) take considerably more time than they once did.

[johnshade]

"Renown" is a noun, not an adjective. The word you are looking for is "renowned."

[SoloSlacker]

The real question is how does BigLaw get away with charging such high rates for junior associates that are just reading and coding emails from home?

My big law buddy gets paid well but he hasn't done anything beside discovery production for over 2 years.

I once asked him a pleading question - he had no idea. When he asked his senior associate, he had no idea either - the senior associate answered: "why is he asking us?" (I found out later that this senior associate has never been in a court, or taken depo and he is in a lit group)

As a solo slacker in the same two years I have many filed complaints and motions, argued and won in federal court, and so on. I am no expert but I am much further on way to being a real lawyer than many big law associates. And I enjoy the work.

There is no arguing that the folks hired by big law are bright. However, I think Big Law firms are honey pots that keep smart over-achieving lawyers off of the streets, and out of the courtrooms, so I can have an easier time of it out here in the real world.

[Russ]

Let me clarify my question about increased technology and its impact on legal fees, because it seems like responders missed the point.

Because of westlaw and computers, it takes far less time to research, write, edit, and finalize memoranda and pleadings. And yet, the lawyer's time at the office has only increased, as have legal fees. Why is that?

[Corey]

1) Firms take more cases per lawyer now.

2) There are no incentives to work efficiently or avoid duplication on an hours system.

3) More hierarchy creates more churn. Computers don't help you when ten partners and 20 associates are sending revisions to pleadings.

4) e-Discovery means faster review per document, but there is signifcant supervisory overhead and the raw number of documents that get produced goes up ten-fold.

5) Honestly, westlaw isn't God's gift to legal research. For a lot of subject areas I can still beat the computer with a well-indexed treatise. Make a computer that actually reads all the cases it finds for me and we'll talk.

The other thing about computer research, combined with the long-term trend of courts publishing every case, is that you frequently end up finding and reading 20 cases on both sides of even the most trivial issue. I do not believe technology has improved the practice of law, though it has certainly made it more mechanical (and empirical).