Situations such as an entrepreneur overstating a project's value, or a superior choosing to under or overstate the gains from a project to a subordinate are common and may result in acts of deception. In this paper we modify the standard investment game in the economics literature to study the nature of deception. In this game a trustor (investor) can send a given amount of money to a trustee (or investee). The amount received is multiplied by a certain amount, k, and the investee then decides on how to divide the total amount received. In our modified game the information on the multiplier, k, is known only to the investee and she can send a non-binding message to the investor regarding its value. We find that 66% of the investees send false messages with both under and over, statement being observed. Investors are naive and almost half of them believe the message received. We find greater lying when the distribution of the multiplier is unknown by the investors than when they know the distribution. Further, messages make beliefs about the multiplier more pessimistic when the investors know the distribution of the multiplier, while the opposite is true when they do not know the distribution.