The Irreversibility Effect: Necessary Versus Sufficient Conditions
|Presenting Author:||Urvashi Narain (Resources for the Future)|
|Coauthor 1:||Anthony Fisher|
|Coauthor 2:||Michael Hanemann|
It is well established that, faced with a linear net benefit or return function or with an all-or-nothing investment choice, agents are more likely to delay investment if future benefits are uncertain, decisions are irreversible, and there is a possibility of learning about future benefits. This paper explores the question whether this effect, which we call the irreversibility effect, holds if investment is a continuous choice variable and the benefit function is non-linear. The existing literature contains a number of necessary and sufficient conditions for the effect to hold that we show are only sufficient, and not necessary. Furthermore, the existing conditions are hard to relate to the primitives of an economic model. We develop a sufficient condition for the irreversibility effect, which, though restricted to intertemporally separable benefit functions, is easier to relate to the primitives of the economic model.
|Link to paper:||Not available|
|Session / Day / Time||17B / Thursday / 4:30 - 6:00 pm|
Please use your browser's "back" button to get to the previous page.