Well, valuable and marketable are two different questions. Valuable is a subjective judgment, IMO, and kind of depends on the individual. Marketable? Eh, that depends on your training and what you do with it.
The best way to approach graduate study is not to think “What can I do with this degree that I want to get?” but rather “What do I want to do, and what kind of degree do I need to do it?” In other words, start with the career and work backwards rather than getting the degree and working forwards. Frankly, getting another degree simply because you’re not ready to give up being in school is not a good reason to get one. You can learn without being in school - there are lots of jobs that enable and foster lifelong learning beyond just being in the classroom. Moreover, I think someone who goes to get a PhD simply because they like to be in school will be disappointed. PhDs are less coursework-based and more research-based; the classes are the least important part.
If you want to be a researcher or consultant for a federal agency, you can do that with a PhD. (You can also do it with an MA, although a PhD might give you an advantage). However, I wouldn’t get it in criminal justice. Sure, there are always new issues that come up, but how many times have you actually heard of the federal government doing some kind of independent research investigation into it with a team of PhDs? A lot of the studies you hear of are done by university professors. The analysts within the federal government tend to have MAs and BAs with experience. (For example, FBI intelligence analyst positions only require a BA. You’d likely be better off with an MA and 5 years of experience in law enforcement or military than you would be with a PhD in criminal justice.)
A broader discipline - like sociology or political science, or economics - would give you more flexibility. But as you work towards the PhD, you have to stay focused on getting the kind of experience that non-academic employers value. PhD programs will not confer that experience automatically. You have to seek it out - do internships, work as a consultant, make sure you get strong statistical/quantitative skills, do some informational interviews and networking with people who work in the federal government.
Honestly, I think the best way to work as a consultant for a federal agency is to work for a federal agency for some time (at least 5+ years, preferably 10-15+) and then offer your services as a consultant. There are certain idiosyncrasies to the system it helps to know.