What are the benefits of getting a PhD in statistics?

A doctoral degree is a credential signifying largely that you are capable of doing independent research at the highest possible level. It is not, in an of itself, a ticket to working in a particular industry or in a particular location any more than a bachelor's or a master's degree. In fact, PhD holders may have substantially more challenges in those aspects, because the additional qualifications make them unattractive for many positions in conventional businesses and industries. (You are unlikely to find a PhD working in a sales division of a multinational conglomerate, for instance.)

The reason to get a PhD is because you are interested in problem solving and doing original work. If this doesn't describe your motivation, I would recommend against pursuing a graduate degree, because it will be a very long few years of your life which are not guaranteed to achieve the objectives you've laid out.

A PhD in statistics is more flexible and useful that PhDs in some other areas. The usual issue with PhDs one hears about is that one becomes over-qualified for non-academic work once one has a PhD. Additionally, there is a lot of time spent getting it.

However, statistics is intrinsically an applied science, and one that is in big demand across lots of areas, because it can be applied to lots of areas, unlike most academic disciplines. Specific anecdote: I was once told by a Statistics Professor that the head of a clinical trial is required to have a PhD in statistics (by the NHS, possibly). I don't know if this is true, but it sounds like something that is probably true. As he put it, this creates jobs for PhDs.

With computers being used more and more, and lots of data being created that needs to be analysed, new methods need to be invented to handle all this data. This is the kind of quasi-research work which is quite well suited for someone with a PhD.

Areas like data visualization and graphics are quite hot right now. Having a PhD in an area like that will probably not hurt you. See Hadley Wickham's thesis for example.

Of course, it is possible to get a PhD from a Statistics Department without learning any statistics, for example if you write a Probability (Mathematics) thesis. You probably don't want to do that.

My personal experience (I have a Statistics PhD) is that to get an interesting work, even in industry, a PhD is helpful. Much of the work so-called statisticians do is to mindlessly apply standard algorithms from some software package to data using things like SAS, and then package up the (machine produced) results. If you have a functioning brain, you don't want to do that.

BTW, it seems such questions are not on topic at stats.sx, but you could ask people on chat there - perhaps point to this question.