First, for those who may not know, an LLM is a postgraduate law degree--a law master's degree. In the United States it is typically a one-year course of study at a law school. In some programs you focus your studies on a particular area of law, but often you do not have to.
Second, I should note that my comments are geared toward US JD holders or students who are considering LLMs. Many LLM students in the US are foreign students with law degrees from home country institutions. The advantages of LLMs to them are very different from the advantages for US law students. These advantages include gaining exposure to the US legal system (and thus making themselves more marketable in their home countries); establishing US professional references; and being able to sit for the bar and gain admission to practice in some US jurisdictions like New York. But for US lawyers and JD candidates, the calculus is different. The following are some factors I think should be considered in deciding whether an LLM is for you.
Factor #1: An LLM can be a way to "shop yourself up." If you have a JD from a national law school, then hurray for you. That helps make you more marketable. But if you went to a regional school, an LLM is a way to improve your cache on the market. "I have a JD from Regional State University" generally will carry less weight than "I have an LLM from Harvard/UVA/Chicago." Not that regional schools are bad--many offer superb education. But like it or not, school rankings matter. And because LLM students typically are in class with JD candidates, you are proving, in essence, that you can run with the pack at a national law school. That is definitely worth something.
On the other hand, if you went to a regional school and have good job prospects, an LLM may not be worth the time and expense. Most US lawyers, in fact, do not have LLMs.
Factor #2: Less competitive admissions. It may be easier to get admitted to a given law school as an LLM student than as a JD student. Partly that is because there generally are fewer applicants--but I also think it must have something to do with the fact that LLM applicants' GPAs and LSAT scores are irrelevant for the US News and World Report rankings. So while you may not have been accepted to a particular school as a JD student, you might be able to get in and prove your mettle as an LLM candidate.
Factor #3: "Shopping yourself up," continued: national versus regional law schools. The standard wisdom (stated above) is that the more highly ranked the law school is, the better. Generally, that is true. But what if you want to practice in a particular region of the country? In that case, an LLM from a respected regional powerhouse may serve you as well as an LLM from a truly national school--and be far less expensive, too. My alma mater, Northwestern, is a wonderful law school, but it is not cheap. If you want to practice law in the Deep South, an LLM from Alabama, for example--which is a top-tier school according to US News and World Report--might serve you just as well.
Factor #4: LLMs as a way to get established in a new market. This is really a corollary to factor #3, above. If you are from a law school in, say, the Northwest, and you want to move to another area of the country to practice law, an LLM is a way to go about this. You have a year to develop local contacts, obtain local references from professors at the law school, look for a job, and meet fellow law students who will be your practicing peers in that market. And it shows your determination to relocate--since you've already done it. That last factor should not be underestimated, since firms are typically reluctant to hire new associates whom they think will not stick around.
I am not suggesting that pursuing an LLM is the best way to relocate, and it is certainly not the cheapest. But I have seen this approach used successfully.
Factor #5: An LLM can be a segue into a specialty area of law. The traditional wisdom was that an LLM was useful for someone who wanted to practice tax law, since that field is so complex. That remains true, but with the continued hyper-specialization of law practice, there are other fields in which an LLM can be useful as well. International trade law and environmental law come to mind. An LLM gives you a year to focus on courses in such areas of law. In fact, I have known lawyers who, after practicing for a while, used LLMs as a way to retool and recast themselves as specialty lawyers. That can be a bold move to make--especially if you are doing OK already in practice--but if you are dissatisfied with the direction of your career it is a way to break out of the rut.
Factor #6: An LLM can help you transition from practice to teaching. In order to land a tenure-track teaching job at a US law school, you generally need to have another postgraduate degree besides your JD. Take it from someone who just recently went through the process. Some law schools want candidates with JDs and PhDs, but you want at least an LLM or equivalent degree (a master's degree in something else). Non-tenure-track jobs at law schools may be somewhat less competitive, but there too, an LLM can give you a leg up in terms of credentials.
In addition, pursuing an LLM gives you a year out of the work force in which you can write and publish law review articles. Landing a law teaching job virtually requires an established record of scholarly publication, and it is easier to write when you are in an academic environment than when you are practicing law 14 hours a day. Again, take it from me: I have done it both ways.
A word of caution: I cannot over-emphasize how challenging and competitive the law teaching market is. If you are thinking about this as a career move an LLM may be part of your strategy, but it is not the only part, and certainly is no guarantee of success.
* * * *
These are by no means the only factors to consider, and I would be very interested in hearing any thoughts others might have on the matter. Ultimately, I think LLMs can offer enormous potential benefits, but for many the costs will exceed the benefits. The above factors certainly suggest to me that LLMs are most advantageous in fairly narrowly defined circumstances.