The study is comparing apples and oranges. The proper comparison is the rate at which minorities are observed by the police violating the law vs the rate at which others are observed violating the law and then comparing that to the relative rates of stops. Two factors confound the analysis--both related to observation:
1. Your question: do minorities violate the law more frequently than others; and
2. Do police observe minority violations more frequently? This later factor could be just a factor of good policing--more cops in high crime areas mean, on average, a greater proportion of minorities will be observed every day by police. Thus, maybe non-minorities drive around with broken headlights at the same rate as minorities (that is, factor 1 for equipment violations is the same as between minorities and others). But if police are concentrated in high-crime districts, they will observe more violations by minorities than the minority proportion of the population. So if they stop minorities at the same rate as others (per observed violation), the result would be numbers like you see in the study.
It would be interesting to see if the numbers change depending on the race of the stopping officer. That would be an easy number for the legislature to require.
The study in question answers your #1 directly. Evidence is, they do. So your apples/oranges assertion is bogus.
Yes they do. For example, in Washington DC where only about 28% of the cops are white, they have the largest disparity of black/white incarceration. In white conservative areas of the country the the ratio is smaller. Blacks in D.C. are thirty times more likely to be in jail.
POLITICS, IMPRISONMENT AND RACE
Another question regarding this issue: would it be more fair if police scrutiny of crime were relaxed for minorities? In other words, would the black community benefit if they did not have equal protection of the law?