The changing income distribution for lawyers (Average is Over)

As of 2010, a graph of starting salaries looks like this:


As of 1991, it looked like this:


That is from Peter Turchin.  Here is a WSJ article by Ben Casselman on the widening job market gap more generally.


What I find interesting is that the 1991 distribution is still with us: the $30k peak inflation adjusts to $50k and the variance and skew around that peak are also about the same. The difference is the appearance of a new "superstar" distribution in addition to the old one.

My thoughts exactly. The superstars are simply the big city "mega firms" of 1000+ lawyers. Move along, nothing to see here.

For curious minds, a "nothing has happened in 20 years" is really interesting.

About income data availability, perhaps the big earners existed since 1991 but somehow data was not available. I'm surprised there's not a classification among tax, realty, crime, immigration, business lawyers. Who are the superstars? Divorce lawyers?

Finally, Tyler just wanted to joke and share this:

The data collection methodology, collector and institutional data collection actors are essentially identical in 1991 and 2010.

High modal value is the industry standard starting salary for new associates at large law firms disproportionately in financial/legal centers like New York, Boston, Chicago, San Francisco and Washington D.C. The low mode is a mix of government jobs (e.g. deputy district attorneys), small and medium sized law firms, non-profits, etc.

In Big Law firms, there are basically two associate tracks: transactional and litigation, which pay the same to start. A typical transactional law associate will spend vast numbers of hours doing "due diligence" and document review for public offerings of securities. A typical litigation association will spend vast numbers of hours in discovery practice - reviewing disclosed documents from the other side, reviewing documents to be disclosed and preparing privilege logs of documents to be withheld, doing legal research in minor footnote points or long shot theories, and drafting brief procedural motions like requests for extension of time. Typically 1800-2000 billable hours a year (which translates to something like 2400-3000 hours present at work) are expected and billed to clients at ca. $300 per hour for their semi-skilled efforts. To get such a job you need to be either an honors graduate at a top law school (often signified by law review), or one of the top two or three graduates at a lesser ranked school that someone in the firm has a relationships with, and to have "intangibles" that impress in an interview that the applicant will be a "good fit."

2010 is not a typical year, however. It was one of the worst for new hires in Big Law in decades. So the relative size of the the top mode relative to the bottom mode is skewed to the bottom.

Aside from a fifth of the group belonging to a class that didn't exist 20 years ago, nothing to see.

That's not nothing.

Also, a fifth of new law grads aren't getting those jobs, so just what law students are represented in the data?

As I argued below (in implicit support of the original comment), that class almost certainly existed before. If you look at the 1991 graph, the $85K group is big firm NY: the group from $60K to $70K is made up of two groups: big firms outside of NY and mid-sized firms in and out of NY. Most of the associates in those groups now make the same (pre-bonus) as the NY big firm group (some don't have jobs or have much lower paying jobs). If you add up the $60K-85K group, it comes to about 18%, the same as the 2010 $160K group. You can see the effect also in the 2010 graph by the very low numbers from $70K through $160K. David Wright's observation is correct: the important thing to notice about the two graphs is that the top group is about the same size and the variance of that group's income over the median is approximately the same.

Your comment is just way too many words to say "yes, there was a top 18%" without mentioning that there's a huge difference between that group then and now.

Worse than useless.

My point is, there's not a huge difference in the top group. I wasn't saying anything about graduates not getting jobs, which aren't represented on either graph, so it's hard to compare.

I just can't fathom how anyone can look at a situation where there was a basically normal distribution of incomes (lots at the bottom and middle, with some more a bit higher at a rate that spreads the graph) and think that's not hugely different from a graph where the top fifth has no connection at all with the bottom 75%, and it's not even close.

"What I find interesting is that the 1991 distribution is still with us: the $30k peak inflation adjusts to $50k and the variance and skew around that peak are also about the same. "

Law schools have been pumping out more grads than jobs for quite some time, I believe. What's probably happening is a 'floor'; at some point either leaving the profession is preferable, or the lawyer is actually pushed out. If you're earning an *average* of $40K, while running an office and running beaucoup financial risk, one or two bad years probably shuts you down.

I wonder why the peak is _eaxactly_ at 160k and nobody seems to earn 170k or more. 160k includes all earning >= 160k?


I think I recall reading a piece about the fact that over 160.000 a year all sort of extra taxes and conditions kicked in (or exemptions were removed), making people with a salary between 160.000 and around 250.000 effectively earning less than people with 160.000 so there were not many people with salaries in the 160-250k bracket. Alas, with the demise of google reader i can't search for the article anymore.

No. Not saying you didn't read that piece, but it's not true that people making 200k are effectively making less than people making 160. That's just not true--common myth about taxation.

All the "big law" firms use the same lockstep compensation scheme that has started at 160K since I believe 2008. Relatively few are paid this much, and those who do are from a handful of law schools of a certain tier, but those who manage to get these jobs are all paid exactly the same.

That means that as Brad says below these graphs are for starting salaries only.

Interestingly, within the small ecosystem of white shoe firms, these entry level jobs are looked at like internships. The firms bring in waves of grads, who are just glorified clerks. They work them like dogs expecting a fair amount of attrition within the first year. Once they pass the bar and start doing real work for the firm, the next phase is sorting out those who will never make partner. The filtering goes on for a few years, with the drones putting in enormous hours. The wages do rise and they share in the firms bonus pool. The best get on a partner track and are destined for life in the over class. The rest fall down the ladder and land either in minor firms or "go inside" and become a corporate lawyer.

"The best get on a partner track and are destined for life in the over class. The rest fall down the ladder and land either in minor firms or “go inside” and become a corporate lawyer. "

Somebody (on the 'Inside the Law School Scam' blog) had pointed out that this is yet another elephant in the room. Even if you went to a top 10 (probably a top 5) law school and got a Biglaw job, 90% of the time you'll be put on the street within six years. At that point, you join the 90% of your peers who are also on the street. The commenter was saying that the result in NYC was that even grads from the top 5 laws were in glut, since the firms dump 90% of them. You find out in your early 30's that even a Harvard Law degree is nothing special.

I recall Tucker Max, in his infamous story about getting drunk at a charity auction, related how a lawyer in the firm where Max was serving as a summer associate confessed that all of the SF-area law firms had conspired to pay the same rate to all of their summer associates -- a slightly lower rate than the NY/LA law firms. The lawyer further explained that there had been numerous complaints on law student internet forums about the difference in rates, and the price-fixing scheme had fallen apart when one of the firms raised their rates to be competitive with the larger markets. The irony is that Max (apocryphally) had created several user names on the forum/s and had himself manufactured much of the complaining.

The point: it seems that elite law school grads are a commodity. Firms don't know which ones will turn out to be partners and which ones will leave, so they pay them all roughly the same salary.

I graduated from an elite law school in 2008 and work in a big Midwestern firm.

The peak is at exactly 160K because that's the standard market rate for a first year associate at a major firm in New York and maybe a few other cities. The smaller peak at 145K is the standard market rate for firms in most other big cities. Note that this is just the salary, not bonus compensation. There is a lot of "follow the leader" in bonus compensation, but it is more variable across firms than salary.

[There are a very few firms that provide higher starting salaries, and a few unique cases where, for instance, Supreme Court clerks get a much higher than lockstep salary or a large signing bonus.]

The lore at my firm is this: During the dot-com boom, West Coast firms wanted to aggressively compete for the new work coming up in Silicon Valley. But they felt handicapped because the New York firms could tell clients they would get only graduates of elite law schools staffed on their cases. (Only two of the 14 traditionally elite law schools are near the West Coast - there are eight within a few hours drive from New York). So the West Coast firms - particularly Brobeck Phlegler - started bidding up market salaries for first year associates from about $80,000 to around $135,000 over the course of a few years to get Harvard and Yale grads to move to San Francisco.

In the early 2000s, a lot happened. 9/11, the failure of the dot-com bubble, and the collapse of Brobeck when its clients went under. But wages are sticky. No one wanted to cut associate salaries, as that would be viewed as an admission of weakness and would hurt morale. When I applied to law school in 2005, the median graduate of a top school made around $125,000-$135,000. Then there was a boom in New York capital markets that led the New York firms to bid up salaries to $160,000. Salaries stopped there because 2008 hit. Some firms cut salaries, some laid off lots of workers, but many kept the same salary and just slashed bonus compensation. And the legal market has not heated up enough to start off a new bidding war.

These numbers are just salary, and the firms have different bonuses. The second tier firms (about 80 or 100 firms) match the top ten in salary, as a signal that they are players, but not in bonus. So associates at Skadden Arps and DLA Piper (for example) may have the same $160K base, but the Skadden associates will get bigger bonuses in most years, much bigger bonuses in a really good year. Now in a really terrible year, like 2009, it might be that no one gets a bonus. But there will be layoffs at the second 80 firms, whereas the top ten will have paid furloughs.

Top law school graduates are fungible resources, particularly since entry level attorneys don't do work where discernable ability at the hiring stage has any impact on productivity. To a great extent the high end legal market is perfectly competitive since information exchange about hiring here is widely reported in the media. The same is not true at the low end where client ability to pay influences pay as much as competition for talent.

GC's theory is probably wrong. When I graduated from law school there was a similar flat rate of compensation for all top jobs, but it was about $80K. That number has gone up, mostly tracking NYC cost of living, over twenty years.

I think you should note that these are starting salaries.

You should definitely note this, as otherwise the numbers are totally unbelievable - I was ready to reject the numbers as obviously incrorect until I read this comment.

The first line of this post:
As of 2010, a graph of starting salaries looks like this:

One of my law profs said that all he was doing was teaching us to read.

Somehow these graphs seem too neat & convenient. Which makes me suspicious. No lawyer earned 90k + in 1991? No lawyer earns 160k + today? Why is the 160k peak so sharp to rise n fall?

Agreed: I smell a rat. Mind you, it is lawyers we're discussing.

Without even clicking the link I can tell these are just starting salaries. Virtually all of the big NYC firms (and their close competitors from DC/LA/Chicago) pay associates on a lockstep compensation scale starting at $160k; of the few outliers who top that, the highest (pre-bonus) salary I can think of off the top of my head is $180k at a firm in DC.

Partners at those firms can earn more by orders of magnitude.

These are starting salaries for new law school grads. And does not include bonuses (or signing bonuses, which can be substantial for ex-law clerks etc.). Biglaw firms are quite (implicitly) coordinated in their starting salaries for new law school grads, for better or for worse.

I assure you that this is the economic reality. There are lawyers who made more than 90K at a first biglaw job in 1991, or more than 160K at a first biglaw job in 2010, but those are almost exclusively comprised of people whose first law job (and hence starting salary in the charge) is as a modestly paid federal appellate court law clerk, which is what the creme de la creme do when they graduate from Harvard or Yale or Chicago with highest honors as law review editors.

No wonder I passed up law school in 1990 and went back in 2003. I was just waiting for average to be over.

The range is actually consistent: the top end makes approximately 3X the lower mode. I think the difference in the graphs is the nationalization of the top end: where once salaries of big firms differed by region (i.e., Chicago, SF and LA made less than NY), now it's all the same. Thus the spread in the 1991 graph toward the right tail has been lumped into one peak, which accentuates the appearance of a bimodal distribution. If you included bonuses, I suspect the 2010 (or 2012) graph would smooth out on the right tail similar to the 1991 graph. I suspect that the actual distribution of people working at top firms (and thus making signficantly more than their peers) is close to the same as in 1991, with one caveat: another important change is the collapse of the "mid-size" firm in the big markets. In older days, lots of lawyers worked in firms in large markets that didn't pay quite as much as the top firms in those same markets, but more than the median. Again, the effect is to spread the distribution in the 1991 graph as compared to the 2010 graph.

Is it legal for firms to collude while setting starting salaries? A sharp peak at $160k, shouldn't that raise some suspicion if this is indeed happening?

It's not necessarily collusion, more like herd behavior. One firm raises starting salaries and the rest move up to match it so that they aren't disadvantaged in securing elite law school grads. Once at that level, no one dares move down because of the same.

I believe this is called "tacit collusion." Or at least you run the risk that the courts will judge it so...

Then they're probably lucky they're BigLaw.

If someone is paying you $160,000 a year, are you going to sue them for antitrust and risk being a pariah in the legal community and never being able to secure a high paying job at another firm? If you are a lawyer in the antitrust division of the department of justice do you care that some of your friends who make three times as much as you might be the victim of "tacit collusion" in setting their salaries? I think the firms are safe...

"Is it legal for firms to collude while setting starting salaries? A sharp peak at $160k, shouldn’t that raise some suspicion if this is indeed happening? "

It's only illegal if (1) you get caught *and* (2) you get prosecuted *and* (3) you lose the case *and* (4) the penalty is nontrivial.

I speculate that the lawyer salary curve in California will become especially interesting in the next decade. A new California state law specifically allows anyone from anywhere in the world who can pass the California state bar to practice law in California.

And my speculation is that a certain highly populated, relatively poor South Asian country where English is one of the national languages, a country with a lot of experience in the British legal tradition, will begin to supply a significant number of new California lawyers real soon now. All they have to do is show up and start passing the California state bar exam. At first, this may seem unlikely, but I'm talking about a country where a large segment of the population has a solid, hard-core tradition of sitting for and passing notoriously difficult exams.

And of course, this collection of lawyers will likely outsource the bulk of the tedious, boring, legal nitty gritty. Get ready people, legal work in California is about to get a whole lot more affordable.

Get ready people, a new law is about to be repealed.

Hahaha yes, lots of people are going to immigrate to California and take it's notorious bar exam because of the glut of legal jobs that exist in the state. That sounds believable.

And they'll do it as illegal immigrants, I suppose

Actually, this is a trend, and is really based on the interest of large firms which want to bring in foreign lawyers to practice their home country's law in the US while also being admitted to practice in the states.

"Get ready people, legal work in California is about to get a whole lot more affordable. "

Only the larger scale work. It's been pointed out that there's a strong floor on the bottom of the profession. It's hard (to impossible) to practice law as a part-time job, unless your day job is one that you can put off for a week on no notice. If somebody calls you at 5 AM Sunday morning because their kid was picked up on a DUI, you either appear in court Monday morning (sitting there at 8AM until the judge calls your case), or you'll never be called by that person (or their network) again.

If your boss has a problem with you telling them on Sunday that you can't be at work on Monday, and will miss one or two more days in the next two weeks for followups, then you have a choice as to which job you'd like to keep.

Is it that we have an oversupply of lawyers and the high salary peak is the ones getting true lawyer jobs and the low distribution includes those law school grads that did not get a job so switched tracks or ended up as paralegals or in non profits or other such alternative careers?

There were about 3600 of those high paying BigLaw jobs this year and over 40,000 graduates. Which leaves another 15,000+ "true lawyer jobs" in there.

According to the National Law Journal, the percentage of new associates making more than 160k has declined for the last three years, a fact which is not captured by the chart based on 1991 as a comparison.

Furthermore, even comparing the chart of 1991 to today is somewhat misleading for two reasons.

First, the composition of attorneys making initial high salaries is different in both periods. There are far more IP attorneys with advanced technical degrees in the high price segment than there were on the early 90s when IP was not as strong.

Second, starting salaries are "sticker price" salaries. Today, large law firms hire some attorneys at these prices, but they have also created nonpartner starting salary tracks much below the sticker price, which firms like to tout in order to convince you to pay our high fees.

Finally, if you really want to understand how the world really works, understand that firms who start with the high salary are very, very unlikely to make you partner six or seven years later. Some NYC firms take in 60 associates, and maybe only one eventually makes partner.

I noticed that in the 1980's managers and investors discovered that in many professions some people where much more productive than others. Of course Lebron James is still under paid.

This is not an argument for the notion that the average is over, simply that the average is different. Where the proper centrality measure for the earlier distribution is likely the median observation, the most proper centrality measure for the most recent distribution is the mode, in that it appears that this most recent distribution is nearly bi-modal--where a measure for the lower part of the distribution and the upper part of the distribution (two modes) would be the proper approach.

It should probably be noted that the actual distribution of compensation for lawyers is not as starkly bi-modal as the chart makes it appear.

Yes, the 160k first-year base salary has become focal, and for whatever reason firms have found that it is necessary to join the 160k club in order to compete for good law graduates. However, first-year bonuses vary across firms. Also, base salaries start to diverge fairly rapidly after the first year, and there is a good deal of separation in what a 4th year associate makes across firms that otherwise provide identical first-year salaries.

Also, Bill's first and second points above are largely incorrect.

Comments for this post are closed