One of my favorite examples of a prisoner's dilemma is when states compete to lure companies to relocate, each state offering greater and greater development packages and tax breaks. Offering such tax breaks is a dominant strategy—if other states don't offer tax breaks, you will certainly win if you do; if other states offer tax breaks, the only way to stay in the running is to offer them as well.
Earlier this year, Wisconsin offered $3 billion to lure a Foxconn factory to its state. New Jersey is offering up to $7 billion to lure Amazon to open a new headquarters.
The problem is that such offers lead to an arms race in which the tax breaks actually become irrelevant. When every state offers huge tax incentives, companies decide on non-tax factors like an area's labor force, transportation, and quality of life. But that's exactly how companies would decide in the absence of state tax incentives. Thus, the incentives don't change what companies ultimately do, but they sure cost a lot.
As John Oliver observes, that's why some of the states that aggressively offer tax breaks, like Connecticut, see only a return of seven cents on every dollar given away.