As a clinical researcher I wood say the reason for underfunding and lack of research in general is quite simple. It is extraordinarily difficult to think of what one would measure. More or less everything that is obvious to measure has come back normal so far. The problem is just that as a scientific question ME is very hard to see a way in to.
For all the social reasons, I think the biological one set out by Jonathan Edwards is the biggest reason that researchers stay away. It's a really tough problem to crack (though by no means impossible).
I think that NIH Director Francis Collins was right when he said that solving ME/CFS will mean discovering some new system or process in the body: what has gone wrong is probably something we have yet to discover in its healthy, normal form.
On the other hand, if you want to tackle an illness like this, where there are no good leads (nothing specific obviously wrong with people) there are a number of ways in.
1. A GWAS (thanks
@FMMM1 
)
Genetic studies (good ones) can find clues as to what has gone wrong biology based on no hypotheses, no clues whatsoever - other than what is written in the DNA of people with the illness.
2. Good epidemiology. Who gets ill, when and where is a good way to find clues as to why. Again, no biological hypotheses needed.
3. Prospective studies - tracking people through from onset (or, better, from before). That way you can see what predisposes people to get sick, which must be a causal factor (not a chance association, or an effect). In a way, a GWAS is a retrospective prospective study (retrospective in that you wait for people to get ill first, prospective in that the clues were writing in their DNA before they got sick).
4. Challenge studies. Poke people with a stick, and see how they react: are patients different from healthy folk or those with different illnesses? Maximal exercise studies are a good example, though I worry that the test is absurdly extreme (nobody with ME in the real world ever gets PEM because they exercised to exhaustion) and necessarily excludes most of the patient population who are too ill to take it. Even so, it is a powerful method because any differences from controls are likely to be linked to the nature of the illness, not to simply being ill.
2&3 already flagged up by
@Michiel Tack :
I do wonder: why has so little non-biomedical research taken place on ME/CFS, for example on diagnosis, prognosis, risk factors, comorbidities, symptom reports, activity levels etc. It sometimes feels that the team of Leonard Jason at DePaul University, Chicago, is the only one doing this type of research.
Why for example hasn't there been a prevalence study on the European mainland? Given the high disease burden, this would certainly be justified. It could provide answers on many of the things listed above which all don't require a biomedical lead or something to measure.
Another option is to do a prospective study: measure all sorts of things with questionnaires before people get sick and then test if the people who developed ME/CFS had some risk factors that stand out. That would be valuable and doable, even if people don't have good ideas. Judith Rosmalen recently said that she couldn't get funding to get people properly tested for ME/CFS in the Dutch lifelines cohort.
So in short, I think all sorts of valuable ME/CFS research could take place even without hints about the underlying pathology.
AIDS activists were quite aggressive. They disrupted conferences, protested before the FDA headquarters and chained themselves in the office of pharmaceutical companies. They were young men who had received a death sentence and were furious. This doesn't seem to have hampered researchers from stepping into the field.
HIV/Aids is often used as a paradigm, 'here's what we could achieve in ME/CFS if only we had the will and the resources'. I was doing my biochemistry degree in the mid-1980s when things were really kicking off and don't think the analogy holds up. That's because HIV/Aids was clearly a solvable problem. Quickly people knew it was caused by a retrovirus that attacked the immune system. We know very little about retroviruses but molecular biology was at the perfect stage to take on the challenge.
The smartest and most ambitious researchers were desperate to work on HIV/Aids - it was the cutting edge of science, there was plenty of money for it, many researchers knew those that were affected, and there was a whiff of Nobel prize in the air.
Of course, all the political factors have made life harder. As for the claim that BPS researchers have been put off research because of harassment - where is the evidence? They seemed to be doing rather a lot of it, at least until those pesky patients started pointing out how flaky their work was.
