Is Law School Salary Really Bimodal? Forum

[RPK34]

Everyone always brings up how law school salaries, at least in the first year after graduating, are really bimodal, pointing to this:

--ImageRemoved--

From this, everyone generally points out that if you miss the 160k, then you're pretty much stuck in 35k-60k range. But this really isn't true. In fact if you integrate the area under the curve, it actually looks very much different. In reality, a little under half of the graduating class working in firms will make a salary that is NOT in one of those salary ranges

Here's a rough breakdown of how the numbers look:

0-55,000: 35.2%
60,000-100,000: 28.7%
105,000-155,000: 14.1%
160,000+: 23%

(By the way, if you add these numbers up, it comes up to 100.9%. This wasn't an oversight, its toughing dealing without the raw data so there is a small margin of error).

There are a few caveats worth throwing out here:
1. This is firm data reported to the NALP (which, from what I've gathered, will not have reporting bias issues since it is reported by firms in the NALP directory). This means that those who are unemployed, working in retail for 8.50/hr, clerking, etc. are not counted in this data
2. The 28.7% of those making 60k-100k is slightling skewed towards the bottom of that pile. But still, just over 14% of those working in firms make between 75k-100k.
3. The 105k-155k is still probably largely dependent on market. For instance, there's a noticeable spike in the graph at 145k (see graph), which I believe is the biglaw salary in a few secondary markets. Unless you're going to school in one of these markets or grew up there, you're probably shut out of them.
4. Finally, just to reiterate, this is not the salary distribution of law school students. This is data for those who make it into a law firm in the NALP directory in their first year. This is more to suggest that those who talk about the "mythical 85k law firm" as if only a small percentage of students who get into firms are able to get these jobs.

[Renzo]

Wait, you put up a graph showing a bimodal distribution, and calculate that over half of the data points fall within the clusters around the two modes, and conclude that it's misleading? Or did I miss something? Yes, the area under the curve is what matters, but that doesn't really change the takeaway: using "average" salary data is tremendously misleading.

[thelaststraw05]

Everyone always brings up how law school salaries, at least in the first year after graduating, are really bimodal, pointing to this:

--ImageRemoved--

From this, everyone generally points out that if you miss the 160k, then you're pretty much stuck in 35k-60k range. But this really isn't true. In fact if you integrate the area under the curve, it actually looks very much different. In reality, a little under half of the graduating class working in firms will make a salary that is NOT in one of those salary ranges

Here's a rough breakdown of how the numbers look:

0-55,000: 35.2%
60,000-100,000: 28.7%
105,000-155,000: 14.1%
160,000+: 23%

(By the way, if you add these numbers up, it comes up to 100.9%. This wasn't an oversight, its toughing dealing without the raw data so there is a small margin of error).

There are a few caveats worth throwing out here:
1. This is firm data reported to the NALP (which, from what I've gathered, will not have reporting bias issues since it is reported by firms in the NALP directory). This means that those who are unemployed, working in retail for 8.50/hr, clerking, etc. are not counted in this data
2. The 28.7% of those making 60k-100k is slightling skewed towards the bottom of that pile. But still, just over 14% of those working in firms make between 75k-100k.
3. The 105k-155k is still probably largely dependent on market. For instance, there's a noticeable spike in the graph at 145k (see graph), which I believe is the biglaw salary in a few secondary markets. Unless you're going to school in one of these markets or grew up there, you're probably shut out of them.
4. Finally, just to reiterate, this is not the salary distribution of law school students. This is data for those who make it into a law firm in the NALP directory in their first year. This is more to suggest that those who talk about the "mythical 85k law firm" as if only a small percentage of students who get into firms are able to get these jobs.

I'm not sure if you broke the data down or borrowed this from someone else who did, but you didn't disprove in the slightest that it was bi-modal.

I think it would be more helpful to calculate:

0-35k
40k-60k
65k-100k
100k-155k
160k
165k+

The two modes that people point to are the 40k-60k areas and the 160k number. By including 60k in the 60k-100k group rather than in the lower group you are inflating the size of the 60k-100k group.

[RPK34]

Wait, you put up a graph showing a bimodal distribution, and calculate that over half of the data points fall within the clusters around the two modes, and conclude that it's misleading? Or did I miss something? Yes, the area under the curve is what matters, but that doesn't really change the takeaway: using "average" salary data is tremendously misleading.

How is this using "average salary data?" And bimodal, by definition, is having two distinct "forms." In this sense, it would be implying that there are really only two salary ranges for law students (the 35-60k vs 160k). A lot, almost half, will fall in between these two ranges, thus not bimodal.

[rman1201]

Wait, you put up a graph showing a bimodal distribution, and calculate that over half of the data points fall within the clusters around the two modes, and conclude that it's misleading? Or did I miss something? Yes, the area under the curve is what matters, but that doesn't really change the takeaway: using "average" salary data is tremendously misleading.

How is this using "average salary data?" And bimodal, by definition, is having two distinct "forms." In this sense, it would be implying that there are really only two salary ranges for law students (the 35-60k vs 160k). A lot, almost half will fall in between these two ranges, thus not bimodal.

I think he meant looking at the median of 72k and expecting to make that much is whats misleading, since a minority actually end up in the 70-80k range.

[RPK34]

Wait, you put up a graph showing a bimodal distribution, and calculate that over half of the data points fall within the clusters around the two modes, and conclude that it's misleading? Or did I miss something? Yes, the area under the curve is what matters, but that doesn't really change the takeaway: using "average" salary data is tremendously misleading.

How is this using "average salary data?" And bimodal, by definition, is having two distinct "forms." In this sense, it would be implying that there are really only two salary ranges for law students (the 35-60k vs 160k). A lot, almost half will fall in between these two ranges, thus not bimodal.

I think he meant looking at the median of 72k and expecting to make that much is whats misleading, since a minority actually end up in the 70-80k range.

I'm not trying to argue that the median isn't slightly misleading. But what's not immediately apparent from this graph is that there is almost an equal number of firm positions hiring at 65-100k as there are big law positions. Or that there are more positions available in firms paying 65k-100k than there are firms hiring in the 45k-60k salary range.

The data is much more interesting when you look at it in the aggregate.

[androstan]

Everyone always brings up how law school salaries, at least in the first year after graduating, are really bimodal, pointing to this:

--ImageRemoved--

From this, everyone generally points out that if you miss the 160k, then you're pretty much stuck in 35k-60k range. But this really isn't true. In fact if you integrate the area under the curve, it actually looks very much different. In reality, a little under half of the graduating class working in firms will make a salary that is NOT in one of those salary ranges

Here's a rough breakdown of how the numbers look:

0-55,000: 35.2%
60,000-100,000: 28.7%
105,000-155,000: 14.1%
160,000+: 23%

(By the way, if you add these numbers up, it comes up to 100.9%. This wasn't an oversight, its toughing dealing without the raw data so there is a small margin of error).

There are a few caveats worth throwing out here:
1. This is firm data reported to the NALP (which, from what I've gathered, will not have reporting bias issues since it is reported by firms in the NALP directory). This means that those who are unemployed, working in retail for 8.50/hr, clerking, etc. are not counted in this data
2. The 28.7% of those making 60k-100k is slightling skewed towards the bottom of that pile. But still, just over 14% of those working in firms make between 75k-100k.
3. The 105k-155k is still probably largely dependent on market. For instance, there's a noticeable spike in the graph at 145k (see graph), which I believe is the biglaw salary in a few secondary markets. Unless you're going to school in one of these markets or grew up there, you're probably shut out of them.
4. Finally, just to reiterate, this is not the salary distribution of law school students. This is data for those who make it into a law firm in the NALP directory in their first year. This is more to suggest that those who talk about the "mythical 85k law firm" as if only a small percentage of students who get into firms are able to get these jobs.

the number of new lawyers making 90-150 is about 50% more than the number of new lawyers making 155-165. Of course there are also about 20% more making 65-90 than there are making 155-165. And finally, there are about five times as many making 35-65 as there are making 155-165.

155-165: 11.18%
90-150: 16.76%
65-90: 13.41%
35-65: 58.67%

These were obtained by numerically integrating the bimodal curve published by NALP.

http://www.top-law-schools.com/forums/v ... l#p3362571

[T6Hopeful]

There's also the need to consider what market you're in. While the nominal income may be incredibly different in say, NYC vs. Oklahoma, the real income (not accounting for loan repayment, which is still a very real consideration) could very well be equal - so somebody on that graph that's making $90k could very well be the top of their market's bimodal distribution.

Not to say that salaries in the legal field are generally bimodal and that's that, but I figured that was worth bringing up.

[Anonymous Loser]

4. Finally, just to reiterate, this is not the salary distribution of law school students. This is data for those who make it into a law firm in the NALP directory in their first year. This is more to suggest that those who talk about the "mythical 85k law firm" as if only a small percentage of students who get into firms are able to get these jobs.

My understanding was that this data represented all employer types. Also, there are far, far more individual employers represented in the study than there are NALP members.

. This is firm data reported to the NALP (which, from what I've gathered, will not have reporting bias issues since it is reported by firms in the NALP directory). This means that those who are unemployed, working in retail for 8.50/hr, clerking, etc. are not counted in this data

I didn't see any discussion of methodology on the NALP site: how do you know this isn't student- or school-reported data? I'm not trying to start an argument; I just haven't been able to find much information about how NALP structures these studies.


That being said, I do agree with you to a large extent: the TLS conventional wisdom which holds that no legal jobs exist that pay between $45k and $120k is ridiculous and entirely unsupported by the available research. Although purely anecdotal, all but 3 of the ~40 jobs I have interviewed for over the course of my law school career offered stating salaries between $75-$120. Granted, I was looking at boutique firms in secondary markets, but the point is that in my experience, these jobs are not by any stretch of the imagination a rarity.

[RPK34]

4. Finally, just to reiterate, this is not the salary distribution of law school students. This is data for those who make it into a law firm in the NALP directory in their first year. This is more to suggest that those who talk about the "mythical 85k law firm" as if only a small percentage of students who get into firms are able to get these jobs.

My understanding was that this data represented all employer types. Also, there are far, far more individual employers represented in the study than there are NALP members.

. This is firm data reported to the NALP (which, from what I've gathered, will not have reporting bias issues since it is reported by firms in the NALP directory). This means that those who are unemployed, working in retail for 8.50/hr, clerking, etc. are not counted in this data

I didn't see any discussion of methodology on the NALP site: how do you know this isn't student- or school-reported data? I'm not trying to start an argument; I just haven't been able to find much information about how NALP structures these studies.

Possibly you're correct. I had found something on google that suggested otherwise, but I'm not sure what I did with it. I'd be interested in figuring out exactly what their methodology is.

[Renzo]

Possibly you're correct. I had found something on google that suggested otherwise, but I'm not sure what I did with it. I'd be interested in figuring out exactly what their methodology is.

NALP collects this data from employers, not schools or students, so it's fairly accurate as to NALP firms (that last bit is a big caveat, however).

As for whether this distribution is bimodal, I cannot fathom your argument as to why it isn't. By definition it is: it has two local maxima. As for your other conclusions, all those flow naturally from the data. Yes, there are jobs between the two big salary clusters, and yes, it's a non-trivial amount. Congratulations, you just proved that firms 100-250 in the NLJ 250 exist.

[20160810]

I will be working for a mid-sized firm with a starting salary in the middle of the graph. Unicorns exist, my friends!