There are growing inequalities in health and wealth among Americans, a gap between "haves" and "have-nots" that has become obvious as the American middle class has been decimated in the ongoing recession, but that gap is no different in academic science. More and more academic scientists are competing for a pool of money that is not growing as fast as the pool of PhDs who want to stay in academia is.

While scientists in academia are overwhelmingly liberal, when it comes to science funding they are fans of capitalism - the best researchers should receive the most money and if universities acquire those best people, and that means $1 billion per year in NIH funding for Johns Hopkins, they have earned it.

But economic inequality may be hurting science, argues University of Michigan sociologist Yu Xie, the same way it hurts the poor in all aspects of life.  Science funding is much more unequally distributed than other outcomes of well-being such as education, earnings or health. 

The rich get richer, says Xie, a professor of sociology, statistics and public policy, with eminent scientists receiving disproportionately greater recognition and rewards than lesser-known scientists for comparable contributions.  "As a result, a talented few can parlay early successes into resources for future successes, accumulating advantages over time." 

Gini for resource inequality across U.S. universities, 1900–2010. Data reflect total research expenditure, federal research expenditure, and endowment for U.S. universities. Data from Center for Measuring University Performance, which draws from National Science Foundation (research expenditures) and National Association of College and University Business Officers (endowments). 
DOI: 10.1126/science.1252743

While the academic establishment defends these inequalities in a variety of ways, Xie observes that in the long run, resources and rewards must be allocated so that inequality is properly managed and controlled.

Citation: Yu Xie, '“Undemocracy”: inequalities in science', Science 23 May 2014: Vol. 344 no. 6186 pp. 809-810  DOI: 10.1126/science.1252743