Heterogeneity of Investors and Asset Pricing in a Risk-Value World



These conditions are related to those obtained by Leland (1980) in an EU
equilibrium.

4.2 HARA-Based Risk Functions

In this section, we derive more specific results assuming that the risk function
F belongs to the negative HARA-class. F(e) is given by equation (2) with
7 > 2 or 7 < 1. 7 is assumed to be the same for all investors.

A necessary condition for the existence of an equilibrium in the case of a
non-exponential risk function (7 > 
-∞) is

(A+J-_) a+
i                              1


ê£


> о,


V ε with A ≡ ɪɪ Ai and eε ≡ ɪɪ e.


This condition must hold because Ai + e /(1 7) > 0 Vi, ε is required
by the FOC (11) resp. (14). Let e
i ei, then eε = ε e. Hence an
equilibrium requires A + (ε
e) / (1 7) > 0. For the more important case
7 < 1 this implies ε
e A(1 7), V ε. If Wq ≡ ∑i Wqi, then in equilibrium
Wq is the forward market value of the aggregate payoff ε, i.e. Wq = U(επ(ε)).

We first show that investor i’s sharing rule relative to that of investor j
is either concave, linear or convex.

Proposition 4 ; Consider two investors i and j who measure risk by a neg-
ative HARA-function with
7 being the same for both. Then the following
statements are equivalent;

- Investor i,s sharing rule is strictly convex [linear] [strictly concave] rel-
ative to that of investor j.

- Investor i,s sharing constant is smaller than [equal to][greater than] that
of investor j.

Proof. See Appendix D.

22



More intriguing information

1. Optimal Tax Policy when Firms are Internationally Mobile
2. Geography, Health, and Demo-Economic Development
3. Agricultural Policy as a Social Engineering Tool
4. Om Økonomi, matematik og videnskabelighed - et bud på provokation
5. Surveying the welfare state: challenges, policy development and causes of resilience
6. The Role of Immigration in Sustaining the Social Security System: A Political Economy Approach
7. Using Surveys Effectively: What are Impact Surveys?
8. School Effectiveness in Developing Countries - A Summary of the Research Evidence
9. The name is absent
10. Update to a program for saving a model fit as a dataset