When we last left him Professor Campos was tripping all over himself trying to sort out medians and means. Now, however, I think we finally have a set of straightforward testable propositions from him.
Hurray! Let’s open that envelope and read away, Johnny.
Waiting, waiting. Aha. Professor Campos has settled on using lawyer salary medians! An excellent choice.
But wait a minute, there is a problem. Oh no, Johnny tells us Professor Campos hasn’t told us which data base he wants to use. Well maybe that’s because the last time he mixed up data bases (relying on the BLS OES data for lawyers’ incomes and SSA’s data for “all workers.”)
Well, we’re in a generous mood so we will check two of the major databases out there to see what the picture looks like knowing full well Prof. Campos will try to change the subject or terms of debate when he realizes, once again, how far he is from describing the economic prospects of lawyers accurately.(Here is a very helpful primer, from someone much more thoroughly steeped in the data and literature than I am about the importance of choosing data bases carefully.)
Here are the three propositions from our friend in Boulder, Colorado:
“(1) Median lawyer SALARIES have been essentially flat since the mid-1990s.
“(2) Median lawyer SALARIES have grown about half as fast as the average salaries of American workers in general
“(3) Lawyer EARNINGS, including those of self-employed as well as salaried lawyers, have declined markedly since the mid-1990s”
(He makes a fourth personal attack on me that I will discuss briefly below to the extent it is relevant – and it’s not very.)
Those 3 propositions are consistent with the argument made by journalist Noam Scheiber who argued everywhere he could (apparently he wasn’t satisfied with what he said in his original story and took to Twitter and Facebook to defend himself) that lawyers today are worse off than, well, “before.” In fact, Scheiber was even willing to acknowledge he had relied on Campos for help with his article despite Campos’ widely known incapacity to understand higher education economics.
Mike Simkovic has pretty much demolished this viewpoint and noted in his most recent exhaustive response to Scheiber:
“The data shows that after controlling for demographics and inflation, lawyers are earning more in recent years than they were earning decades ago. You are not responding to my point.”
Now, if Mike is right then both Scheiber and Campos are wrong.
We won’t be able to go back decades here but we can see how lawyers are doing over the past fifteen years.
So let’s look at Campos’ latest propositions more closely. Originally he tried to compare lawyer incomes with “all workers” incomes but used one data base for the lawyers and another for “all workers.”
There was no rhyme or reason to this odd mix. And I pointed that out and then Campos went and, well, jumped the shark, sinking to a new defamatory low, as even some of the lemmings who follow him on a blog that Glen Greenwald calls a “cesspool” suggested.
He could have avoided all that nonsense if he had realized that, in fact, the data base from which he took his lawyer earnings numbers also has an “all occupations” category. Of course, as we will see, the results of using an “apples to apples” comparison is not very friendly to Campos’ well known bias against his own chosen profession.
We can make a more accurate comparison if we rely on the same data base (the BLS OES data) for both employment categories.
And what do we find?
Well, let’s go back as far as we can and check the medians.
(We can’t easily go back earlier than 2001. Since Campos does not identify his data base (why not? It’s standard academic practice.) we have no clue as to how he reached his conclusions.)
Median Annual Income 2001/2015
All Occupations $27,060/$36,192
Now those numbers have to be adjusted for inflation since the data collected by the BLS is actual dollars paid at the time.
A lawyer paid $88,760 in 2001 would have $118,789 worth of buying power in 2015. So that means lawyers have fallen behind slightly since 2001. But, similarly, “all occupations” have fallen behind, too. At the median of $27,060 they would have $36,215 in buying power in 2015. So we know the general wisdom that the middle class is falling behind has some validity. Go Bernie.
Clearly Campos is wrong about the medians – they have increased substantially and they have (just) kept up with inflation. Notice that lawyers are not any worse off than anyone else. Despite what the scam bloggers think/imagine/hope/pray happened to the legal profession, there is zero evidence that somehow lawyers are falling precipitously into a lower socio-economic class. In fact, they are holding their own quite nicely.
And since lawyer salaries are high and their salaries are included in the “All Occupations” category that category actually has to have a slightly lower value if they were excluded (hard to do with the data available). In other words, Campos isn’t really telling us anything useful at this point other than that lawyers are in the same boat as everyone else.
But what about his other two conclusions? These would be more significant, if accurate.
He says median lawyer salaries have grown at half the rate of American workers generally. But what do the data above show?
Growth rate of lawyer salaries:
Lawyers have increased their nominal non-inflation adjusted salaries by $27,060 or 30.5%.
Workers generally (All Occupations in BLS OES terms) have increased annual salaries by $9,132 or 33.7%, also non-inflation adjusted.
So “all occupations” (again biased upward slightly because higher paid lawyers are included) have grown by 3% more than lawyers (about a ten percent differential in the growth rate).
Of course, we can use the inflation adjusted numbers, too. These show that both categories of workers have fallen behind slightly. Lawyers lost 2.4% due to falling behind inflation since 2001 but of course their absolute incomes are 3x those in the “All Occupations” category so it hurts a bit less at that altitude.
“All occupations” lost only a de minimis amount, they are flat (again keeping in mind lawyers are included in the “All” category so the actual loss would be slightly higher.) Not much difference in either case.
So on proposition #2 Campos is wrong – instead of a 50% differential in the growth rate there is only a 10% differential (at the most because again lawyers are counted in the second category).
And finally we get to #3 where Campos says (all) lawyer earnings have fallen significantly since the mid-1990s. Here we also can weave in Campos fourth snide personal remark about my suggestion that there is only a relatively small number of solo practitioners out there. Campos thinks the number is quite large and that solos have taken a huge hit since the credit crisis. And that helps him, arguably, get to the conclusion that lawyer earnings have “declined markedly” since the mid-1990s.
Campos criticizes my suggestion in an earlier post here that only 4% of all lawyers are solo practitioners. I took that number from the NALP database that tracks recent graduates and it has a widely understood limitation because it is based on recent graduates (in other words if 4% go into practice as a solo each year their overall weight in the profession would be somewhat larger; this didn’t matter much one way or the other for that earlier post which was testing another proposition about lawyer employment).
The NALP data therefore differs from the data that the ABA has published which is where it appears Campos comes up with a much larger figure. That ABA data, however, is in turn derived from data compiled by the American Bar Foundation based on data from 2005 so it’s a decade old and there is little definition of the term provided.
So we really don’t have a clear answer from that data about the number of solos. But the actual number or percentage turns out to be not all that important for understanding the economic circumstances of lawyers and we will have a fix for this problem in a second, so hold on.
There is another complexity because the BLS OES data above – which show workers and lawyers both staying essentially even with inflation over the last fifteen years – does not include self-employed in “unincorporated firms.” Many solos, however, appear to form separate business organizations that may be picked up in the BLS data.
Data (up to 2014) provided by the American Community Survey that is carried out by the US Census Bureau includes what it calls “self-employed” lawyers, both “incorporated” in separate business entities and “unincorporated.” [Thanks to Mike Simkovic for generating this data.]
The first category likely could be a reasonable proxy for solo practitioners (putting them at about 12% of the national total of the 1 million lawyers the ACS thinks are out there; anyone using the ABA/ABF data would end up with a higher number since the ABA data says there are 1.3 mn lawyers). But even though that larger number might give Campos some comfort, it remains uncertain since self-employed unincorporated would appear to include partnerships. (Partnerships can be viewed as “entities” but they are not incorporated.)
The second category – “self-employed incorporated” – appears to earn a substantially higher income and so it is not clear that these really are solo practitioners.
In any case, the key is what lawyers earn, whether or not in solo practice, and the ACS suggests that solos are doing reasonably well. And so are lawyers generally. In other words, we end up with data that is broadly consistent with the BLS OES data.
Let’s break it down.
First, unlike the BLS OES data we only have means here not medians. Usual caveats apply (even Campos likely understands these by now). But the point is to see the overall trend.
Mean Annual Income 2001/2014
(Private sector) $134,734/$186,375
Now, using the BLS CPI calculator let’s see what happened to the buying power of lawyer earnings over this time period.
“Self-employed unincorporated” (our proxy for solo practitioners) – buying power of 2001 salary in 2014 = $155,693. That means they have gained an additional $9,134 (or 5.8% bump) in buying power relative to where they were in 2001.
“Self-employed incorporated” – buying power of 2001 salary in 2014 = $202,963. Here we see a loss in buying power of $16,718 (or a decline of 8.2%). It should be noted that this category appears to have the greatest volatility in earnings. For example, in 2013 it was $197,347 which put their buying power at $199,723 for a loss of only $2376 or 1.1%.
“Wage/Salaried private sector” – buying power of 2001 salary in 2014 = $180,103. Here we see a jump in buying power of $6272 (or an increase of 3.4%).
Finally, all lawyers – buying power of 2001 salary in 2014 = $162,892. That’s an increase of $1,055 or a bit less than one percent. Not a lot but it just means they held their own against inflation overall and maintained their status as high income earners over a very volatile fifteen year period.
So overall there is no basis to statement #3 of Campos that lawyers’ earnings “declined markedly.” A big chunk of solo practitioners and all private sector employed lawyers did quite well. Lawyers’ relative place as high earners compared to others in the economy was maintained quite easily. And notice the relative results did not change much when we switched from medians to means.
So in conclusion
1) Are lawyer salaries flat over the last decade and a half? No, they have increased steadily and kept up with inflation.
2) Have they grown at a rate half that of all workers? No, they have lost some ground but far less than that and given the large absolute earnings number for lawyers it is fair to say they have held their own.
3) Have lawyers earnings, when you include a reasonable proxy for solo practitioners, “declined markedly”? No, they have in fact increased substantially in nominal terms and with some exceptions have kept well ahead of inflation.
So back to class for Professor Campos!