What percentage of the people covered by deinstitutionalization consists of men like Laugher?
There is no way of measuring that. The nature of mental illness and how we define and treat it has changed so much that it is difficult to measure who would or would have been institutionalized.
But it is pretty damn low. We've got oh, a dozen or so mass shooters who are clearly mentally ill over a period of a couple of years, against tens of thousands of people who are or have been treated for psychiatric problems on an inpatient basis, many of whom may have been candidates for institutionalization in the past.
The prison problem is multi-faceted.
First, prisons, like any ongoing stressor, can induce or exacerbate mental illness in someone who may be vulnerable. Prison is not exactly the kind of place and lifestyle that is conducive to peace of mind. What would be easily manageable on the outside may be a disabling problem on the inside. What would not be noticed or measured on the outside becomes more obvious on the inside. I've lost count of the number of me who have drifted in and out of my life who have some pretty serious mental health problems that they won't seek treatment for--that is less the case in prison. Fewer incentives for avoiding treatment, perhaps more incentives for seeking treatment, and seriously exacerbating conditions.
Second, it's real easy to scam a mental health provider. If I was in prison, you bet I'd be scamming all the psych meds I could, just to keep awareness as low as possible.
Third, poor people are convicted more and receive longer sentences than financially stable people. Mentally ill people are generally poorer than people without mental illness.
Fourth, mentally ill people are convicted more than non-mentally ill people. Mental illness can look scary.
Fifth, and this is one that bugs the heck out of me personally, we seem to have confused evil and mental illness. Enjoying slicing people up and eating their faces does not a mental illness make, but when reads reports of such things, the immediate response is often "That guy is clearly crazy." No, that guy is clearly evil. Being different in a repulsive way is not necessarily mental illness. Sometimes it's just being evil. Of course, there are bona fide "illnesses" listed in the DSM that generally cause no distress to their "sufferers", do not appear to be treatable via medication or therapy, and the diagnostic criteria mostly consist of various ways in which in the individual is an real bastard. I'm not convinced that those belong in the DSM.
I'm not suggesting that Loughner is not bugnuts insane, but I don't know enough to say that he is, either. I know he talked about hurting people, threatened people, and liked reading crude expressions of politically-based hatred. I don't see how those things indicate mental illness. (Disclaimer: I know some people here have looked at the guy's writing and concluded he was nuts, along the lines of "time is cubed" [thanks, Fistful], and I don't dispute that, I am just saying that enthusiasm about violence and having random thoughts about lashing out at a variety of people is not a mental illness. It probably indicates someone who is a bad person and who does not have many critical thinking skills.
However, I'm not disputing the statement that some people with mental illness are violent and that their violence may be related to delusions or lack of appropriate social skills or self-control, but the problem of mental illness in prisons is about a lot more than deinstitutionalization. I notice that the graph of murder rates show less direct correlation. The murder rate is also a pretty complicated problem, and a lot of whole murders are also drug-related--often a result of criminalization. I would be really interested in seeing a breakdown of what sort of murders those are--gang/drug violence, domestic violence, felony murder incident to burglary or armed robbery, or crazy dude randomly shooting people because he thinks God told him to.
About those graphs: There are a whole lot of other factor that affect the prison population size and its growth since the early 80's, like the War on Drugs and the growing acceptance of expanding and expanding again and expanding again what is a felony and can lead to prison times. When some of the most onerous mandatory minimum laws were repealed, they were replaced by growing numbers of small-time MJ growers/sellers/users being imprisoned, and the number of other non-drug-related felonies growing exponentially.
Further, the vast majority of people in mental institutions during the age of institutions were not violent; I don't doubt that many who were, and who were released or not institutionalized, committed some murders. I don't suppose you have a graph that breaks down institutionalization in mental institutions and prison by sex? That might be interesting.