The profession was much more of a bloodsport. Less cases settled, far less went to arbitration, and more money was spent when a case did go to trial.
Nowadays the culture has shifted almost completely to keeping costs down and settling things outside of court -- either settling outright or through a mediated arbitration (much less expensive).
It's no surprise to me that law school grads are out of work these days.
First, there are too many low quality law schools churning out people who can barely pass the bar. Those schools take home the tuition money and the young graduates have a mountain of non-dischargable debt with nowhere to turn. The rate of unemployed young graduates is actually probably even higher than figures show because schools will pay a small stipend to recent graduates so that they can say they are "employed" when the yearly figures for former students employed 6mo out are released. It's all about taking home the tuition for these schools and they shamefully advertise how demand for lawyers will increase and job prospects will remain high.
Second, the business has become vastly more efficient as firms begin to adopt digital case management systems and can now conduct research online instead of having to sort through volumes of old books. You know when you see the generic shelves of law books in offices? Yeah those just sit there collecting dust now. They are for show because all of that information has now moved onto the computer. The legal industry has been slow and late to adopt technology but it has made lawyers vastly more efficient than they could have been before. Even things like being able to email scans of documents can speed up the process astronomically. As older lawyers who cannot use computers begin to exit the workforce and younger ones who can move up in the ranks, firms just do not need as many people working as they did before. The vast majority of legal work is not what you see on TV or in the Ellen Pao case where it is a battle of the brains in real time for a court room. It is mundane routine paperwork that can be automated or done by assistants.
A lot of this stuff can even be done at home by your average person with products like turbo tax that enable the average person to not need representation for basic needs legal needs anymore.
I no longer plan on going to law school and have been taking online courses in programming/reading a ton of programming books since I graduated in hopes of turning what has been a long term hobby of mine into a career, hopefully combined with politics or law. (I am thinking campaign research/advertising online or legal software on the law side.)
In my experience, this is 100% true. I spent a few years working in "legal technology", which is just industry parlance for categorizing documents into "databases" (really just proprietary applications that compete over how best to read data from a flat file).
At the time (2009 - 2011), a ton of work was being done by teams of contract review attorneys. Entire rooms would be filled to the brim with these guys, paid anywhere between $20-$40 an hour to tab through page after page of document metadata and images.
The entire industry was on the verge of being completely upended, and this is in fact starting to have effect. The article on topic modelling and LDA that's also on the front page today is particularly germane, as there are in fact implementations of topic modelling and LDA for the legal technology world.
Look at this old web site: www.shitlawjobs.com
The "demand" (number of disputes) increases with the number of interactions, which is roughly the square of the number of people (each pair is another potential dispute). So if
- population is increasing, and
- supply (courts and lawyers) does not increase as fast as the square, and
- legal technology does not increase output per person[1],
then the backlog steadily increases and people have to work around it.
[1] Many economists insist that law inherently experiences Baumol's cost disease of resistance to technological improvements due to high human component of output, but I think that's more an artifact of the constrains we place on courts.
It wouldn't actually be a square of the population. There's a fairly low number of interactions a single person can have in a day. Plus most disputes that I can have with a person I don't know are criminal instead of civil.
And you have potential interactions with anyone within a given jurisdiction, the same one over which the infrastructure exists. Sure, you don't interact with them all, but you interact with a fraction of them, so the scaling properties are the same.