The flexibility and spending hypotheses predict opposite relations between the control of agency conflicts
and cash reserves. The flexibility hypothesis predicts that poorly controlled managers will be observed to have
larger cash reserves and the spending hypothesis predicts they will have smaller cash reserves. As we noted in
the introduction, which one is correct depends on managers’ tradeoff between current overinvestment versus
future flexibility and the probability of discipline associated with each alternative. The shareholder power
hypothesis predicts that there will be a negative relation between agency problems and cash reserves, similar to
the spending hypothesis, but its prediction is primarily driven by better-controlled managers holding larger
cash reserves rather than worse-controlled managers holding smaller reserves