Date of Degree


Document Type


Degree Name





Christos Giannikos

Committee Members

Sebastiano Manzan

Barry Ma

Subject Categories

Economics | Finance | Income Distribution | Insurance | Other Applied Mathematics


Multivariate Risk Aversion, Insurance, Criteria for Decision-Making under Risk and Uncertainty, Inequality, Bidding-Auctions, Financial Economics


All orders of risk attitude have been extensively studied within a univariate utility framework. For multivariate utilities, much still remains to be addressed, especially regarding the intensity of risk attitudes. This dissertation consists of five chapters and contributes to our understanding of the market participants’ behavior and the generalization of well-known results regarding risk attitudes.

Chapter 1 is a study of decision problems under two-dimensional risk. We use an existing index of absolute correlation aversion to classify bivariate preferences conveniently with respect to attitudes toward this risk. This classification appears to be more important than whether decision makers are more correlation-averse or correlation-seeking. Based on the comparative statics of a standard insurance model for changes in correlation aversion, we derive very simple testable implications about the specification of utility functions. Finally, for the particular case when the two-dimensional risk can be interpreted as risk on income and health, we identify preferences that best fit the observed behavior.

Chapter 2 explores factors that affect disability insurance (DI) ownership and emphasizes how health status and risk tolerance affect the demand for DI. A plethora of individual’s characteristics play a key role in the decision for DI ownership. However, findings from empirical studies of the Survey of Consumer Finances (SCF) indicate that the level of risk tolerance is not a significant factor. Our model is enriched with a new variable introduced in the 2016 SCF that acts as a proxy for risk tolerance and significantly affects the demand for DI. The important role of these empirical results is highlighted by employing recently introduced indices for multivariate risk attitudes.

Chapter 3 examines how to measure absolute bivariate risk aversion within the framework of decision problems under two-dimensional risk. We demonstrate the important role of the univariate risk aversion coefficient on the ranking of preferences with regard to absolute bivariate risk aversion, also known as correlation aversion. Finally, we introduce a new index that correctly captures the effect of univariate risk aversion on correlation aversion.

Chapter 4 generalizes the index of bivariate risk aversion for preferences with multiple attributes. We propose how the framework of Atkinson about inequality aversion can be generalized by using the new measure of multivariate risk aversion. Thus, we introduce multidimensional inequality aversion as a characterization for a welfare function with more than one attribute. A generalization of the most popular welfare-based measure of inequality, the Atkinson Index, is illustrated based on Russian household data between 1995 and 2015.

Chapter 5 focuses on the market mechanism of the blind principal bid (BPB). This study extends the current literature on BPB, in which asset managers auction a large basket of stocks to brokers who know the general characteristics of the basket but not the actual names. The ownership of the basket is transferred to the brokers, who have to complete all the ordered transactions within a short period of time. The commission is paid on the basis of the winning bid at the auction, usually submitted as cents per share. This article provides a simple model that may help brokers formulate their bids. The model’s results are then compared to the winning bids from a unique dataset of 140 baskets. Since the brokers face multiple risk factors that affect future income, their behavior in these successful auctions is discussed.

This work is embargoed and will be available for download on Wednesday, February 01, 2023

Graduate Center users:
To read this work, log in to your GC ILL account and place a thesis request.

Non-GC Users:
See the GC’s lending policies to learn more.