Frederik Lundtofte

The Effect of Information Quality on Optimal Portfolio Choice

Three types of agents acting on different information sets are considered: fully informed agents, insiders, and outsiders. Differences in information quality are shown to affect the properties of their optimal portfolios. For an outsider, the share of wealth invested in the stock is decreasing in the variance of the stock. However, for an insider, the effect of an increasing stock variance on the optimal portfolio weight is ambiguous. In a calibration to U.S. data, the confidence intervals of the insiders demand for the stock converge, whereas the outsiders confidence intervals become wider.

The Quality of Public Information and The Term Structure of Interest Rates

This paper analyzes the term structure of interest rates in an exchange-only Lucas (Econometrica 46:1429–1445, 1978) economy where consumers learn about a stochastic growth rate through observations of the endowment process and an external public signal. We allow for deluded consumers, who exaggerate the degree of covariation between the external public signal and the growth rate. With such consumers, there can be a premium for noisy external public information in long-term bonds and the social value of more precise public information can be negative. Moreover, our model can create excessive yield volatility and deviations from the expectations hypothesis.

Endogenous Acquisition of Information and the Equity Home Bias

This paper investigates the extent to which differences in information costs can explain the equity home bias puzzle. In a model where information costs are higher for the Foreign asset than for the Home asset, I show that, if cost functions are convex and the assets have identical return characteristics, the expected size of the home bias in terms of differences in expected demands is positive and increasing in expected excess returns and risk, but decreasing in risk aversion. However, a calibration to US data suggests that information costs can explain only a small fraction of the observed home bias.

Implied Volatility and Risk Aversion in a Simple Model with Uncertain Growth

We show that a simple equilibrium model with uncertain growth is able to simultaneously generate patterns in implied volatility and risk aversion that are similar to the ones observed in the data.

Risk Aversion, Noise and Optimal Investments

In contrast to the efficient market hypothesis (EMH), the noisy market hypothesis (NMH) asserts that prices are but noisy indications of fundamental values. The authors study losses in certainty equivalents of investing according to one hypothesis (NMH or EMH) when the other is true. Their findings suggest that, for reasonable parameter values, investing according to the EMH when the NMH is true yields lower losses than investing according to the NMH when the EMH is true. Further, investing according to the right hypothesis is much more important for risk-tolerant investors than for risk-averse investors.

Loan Pricing over the Business Cycle

Based on a screening model, we hypothesize that borrower risk will be over- (under-)priced in recessions (booms), and the loan spreads’ sensitivity to default risk as a function of economic growth will be inverse U-shaped. We test this prediction using a sample of 5,300 U.S. commercial loans contracted between 1987 and 2012, and find that the data support the prediction. The result is particularly strong at business cycle peaks: in the main results, moving from zero real GDP growth to the 99th percentile of economic growth, the responsiveness of the spread to borrower risk (the “price of risk”) is more than halved. The results are robust to a variety of alternative specifications and competing explanations.

Institutional Quality, Trust and Stock Market Participation: Learning to Forget

We explore the relation between institutional quality, trust and stock-market participation. In our theoretical model, agents update their beliefs in a Bayesian manner based on observations on frauds and choose whether to invest in the stock market. The corresponding empirical model shows that institutional quality affects trust and that the part of trust that is explained by institutional quality influences stock-market participation. For immigrants, we consider learning factors, such as education and duration of stay, and we find that the impact of the institutional quality of the country of residence, relative to that of the home country, tends to increase with education.