What I have attempted to do in this paper and a companion paper [1] on dividend-signaling is to delineate two “polar cases” of signaling in which firms either can or cannot (at all) directly communicate the ex-post profitability of their business without moral hazard. In this paper, we assume that they can, and signaling through dividends or earnings forecasts “merely” serves to bring forward the timing of communication of insiders' expectation of profitability to the outside market. The model is “nondissipative” because the incentivestructure that leads to self-selection using the signal is based on market value revisions which are themselves based on the discrepancy between the signal and the ex-post indicator, not on any exogenous or “third party” costs, unlike the model in the companion paper [1].

This paper presents a new explanation for the use of debt financing, particularly private debt, in addition to equity without relying on the existence of taxes or bankruptcy costs. The paper assumes that information about returns on investment projects is costly and subject to efficient specialization, so that managers of firms develop inside information not possessed by the market. Suppose the manager of a firm possesses such inside information about a new investment project and his objective is to act in the best interests of existing equity owners. If the information can be disclosed to the market without impairing the value of the project, he will do so. However, much information will be of a strategic nature where the value of the project depends upon confidentiality. Public financing of such projects without disclosing the information will mean that the excess value or surprise monopoly profits in the new project will be split between new and old owners.

The two related problems of agency and informational asymmetry have received increasing attention in finance. In particular, prominent authors in this area (e.g., Jensen and Meckling [7], Ross [15, 16], Leland and Pyle [10], etc.) have demonstrably argued that the financial structure of the firm can be determined in the process of eliminating, or at least reducing, the costs associated with these problems.

In a dynamic economy with a sequence of markets over time, there are generally goods or securities that will be traded in the future at currently unknown prices. Individuals require some notion of what these future prices will be since knowledge of future investment opportunity sets is relevant when making current portfolio allocation decisions.

Sudipto Bhattacharya's paper, “An Exploration of Nondissipative Dividend–Signaling Structures,” is in two parts. The first develops a nondissipative “quota–based” signaling model. Optimal contingent (wage) contracts are shown to induce workers, whose productivity is not directly observable, to “self–select” by the choice of level of committed productivity. However, the most novel feature of the paper is the attempt to apply the quota–based signaling model to the problem of dividend signaling in the capital– –market that is, the modeling of the “information content” of corporate dividends as an ex–ante signal of future earnings. As Bhattacharya notes, the “information content” of hypothesis of dividend payment is somewhat ill–defined to date, so such an application of the quota–based signaling model, if successful, would be a definite contribution to dividend theory.

Haugen and Senbet address several interesting issues, such as: a costless solution, with the use of options, to the agency problem and the incorporation of informational asymmetry into the agency problem, as well as the use of options to convey information. I'll start my discussion on their paper by providing general comments which are followed by some clarifying comments on the authors' argument for signaling with options.

I have two comments to make. Neither is really addressed to the details of the paper. Rather they are prompted by the general line of analysis of information revealed in a rational expectations equilibrium.

I'd like to begin by thanking the Western Finance Association for the lunch I just consumed …

It is only fair that I inform you at the outset that the views you are about to hear can only be described as biased. They are biased because I'll be limiting my remarks to those parts of finance that I think I know something about; secondly, my comments will contain a disproportionate reflection of my own work. The more generous among you might argue that this puts me in good company. A better explanation would recognize that I am really in a monopoly position for the next half hour or so: there are no contemporaneous sessions within commuting distance, your lunch was paid in advance and is not refundable, and for some of you at least there is a certain cost associated with getting up and leaving in full view of the organizers.

Increasing attention has been focused, as of late, on the relatively low rate of rental housing starts and the increase in apartment conversions to condominium ownership. By some estimates, additions to owner–occupied housing stock since 1970 have occurred at twice the rate of addition to the rental stock, a pattern that has caused concern to some policymakers.

The findings of this study indicate, contrary to the recent claims by Aaron [1, 2], among others, that the real estate tax on residential properties is regressive. Using a data base from a sample of household observations for FHA–HUD 203–Program single–family market sales for counties contained in five major United States Standard Metropolitan Statistical Areas (SMSAs), the statistical analyses suggest that the degree of observed regressivity varies significantly across counties and is the result of two forces. First, in many counties, the poor and de facto inequitable assessment practices are the principal causes for property tax regressivity. Second, in some counties that exhibit relatively equitable and uniform assessment practices, the end product appears to be an income regressive property tax. Hence, the property tax in these counties may be intrinsically (slightly) regressive. However, in general, if there were uniform administration of the property tax, it would be (slightly) progressive.

”Poor people cause poor housing.” This statement is the basis for one of the major policy proposals to eradicate poor housing and neighborhoods that plague some of our urban areas. If the motto is correct, then poor housing and poor neighborhoods can be eliminated by providing direct support for the demand for housing by low-income households. The role of government, under such a policy, is to funnel its resources to households in the form of demand subsidies. Examples of such demand-side programs are housing allowances and the Section 8 existing housing program.

This paper presents a hedonic index of residential services fit to 1975 data for the St. Joseph County, Indiana, rental housing market. The work it reports is part of a larger research effort, the Housing Assistance Supply Experiment (HASE), being conducted for the U.S. Department of Housing and Urban Development.

I have always considered it unfair for a discussant to criticize an author for writing a paper differently than he would have written it had he been addressing that topic. However, since the topic of the Brueggeman and Peiser paper is so important and the paper is so different from what I would have written, I am going to indulge myself for a moment.

The Edelstein and Follain papers are not as different as they may appear at first. The two are linked through the income elasticity of housing. Follain is concerned with this measure explicitly, while Edelstein1s analysis hinges on the value he assumes for it. Both papers are also policy papers, and their conclusions are of immediate importance for ongoing policy decisions.

Random stock returns result from irregular vibrations of a share's price through time. Divide any arbitrary time interval into two mutually exclusive and exhaustive sets. One set contains time periods when trading is formally open on an organized market such as the New York Stock Exchange (NYSE). Its complement contains closed trading time periods, i.e., when the NYSE is not open. Conventional theory assumes that the same return process operates over all periods in both sets. No allowance is made for possible differences in the return sequence between sets or among time periods within each set. There are reasons to assume that such differences may exist. For example, during a trading day, stock prices fluctuate as orders are executed. During nights, weekends, holidays, and holiday-weekends there are no transactions, but a share's value from close to open on the next trading day may still change to reflect revised expectations about a firm's productivity. In fact, capital changes and important news items are usually announced after the stock exchanges close.

Academic attention has increasingly been focused on the operation of security markets. This is largely due to the impetus provided by the Institutional Investor Study (see U.S. Securities and Exchange Commission [35]), by the Securities Act Amendments of 1975 whereby Congress mandated the development of a national market system (NMS), and by the expanding computer technology of the 1970s. Not surprisingly, much of the attention has focused on the role of dealers and stock exchange specialists as market makers. The literature has generally viewed these market makers as suppliers of immediacy to ordinary traders, and has taken the bid-ask spread to be the price they impose for the provision of this service.

The first hypothesis underlying this study is that successive transactions exhibit systematic patterns. These patterns will be studied to better understand (1) the processes by which transactions are arranged, (2) the costs of transacting, and (3) the statistical characteristics of the reported transactions prices and their relationship to the market equilibrium. This study will not directly consider market efficiency, though its results may eventually lead to more searching and meaningful studies of market efficiency.

The paper by Cohen, Maier, Schwartz, and Whitcomb (CMSW) presents the authors' comments on previous literature about bid-ask spreads in securities markets and the way in which they are formed and also gives a number of the authors' own ideas on the topic. I have a few comments on what they have to say in both of these categories, but I will emphasize the latter, since I find it more interesting to comment on their own assertions than to comment on their comments about others. In this and previous versions of their paper and previous papers on the same topic, CMSW have explained that they view price movements over time in securities markets as being partly the result of underlying economic changes and partly a reflection of the impact of idiosyncratic orders that come in from individual investors. On the latter point, I do not find their view very convincing. Individuals decide to trade a security, I believe, for essentially one of two reasons: a desire to change their cash balances or a result of a change in probability beliefs about future returns. I do not believe that either of these situations will result in an impact on security price unless they influence to action at the same time large numbers of investors, in which case they represent exactly the source of underlying economic changes which ought to affect price.

One of the interesting anomalies of asset trading in secondary markets is that it is not usually possible to observe actual market clearing prices. Rather, we can only observe transactions as bid and ask price? and infer that the market equilibrium price lies between them. Professor Smidt analyzes continuous transactions data for individual NYSE listed securities during 1977. From it he deduces (1) the behavior of transaction prices, (2) the average size of bid-ask spreads, and (3) the movement of market prices. Given that neither the bid-ask spread nor the actual equilibrium price is observable from the available data, this is an ambitious undertaking. The problem is further complicated by the fact that the transactions data contain three distinct types of trading activity: matching trades at the opening or reopening of the auction market, auction trades (the bulk of activity by number of transactions), and block trades (the second largest activity by dollar volume).

Hedgers are an integral element of most models of futures markets. They are typically viewed as involved in the storage or production process and attempting by futures market transactions to avoid price risk associated with holdings of the underlying commodity. Speculators accept the risk and receive compensation whose size is in considerable dispute. (Keynes [16], Telser [24], Cootner [6], Dusak [8]). This “insurance” view of hedging is sometimes expanded to allow for “discretionary” or “selective” hedging which tends to arise when expectations differ across individuals. Narrow models of hedging in the commodities market (Johnson [14], Heifner [11], Peck [19]), in the foreign exchange market (Ethier [9]), and in the bank loan market (Pyle [20]) as well as more general models of the determination of spot and futures prices that incorporate hedging (Stein [22]) have preceded or ignored the theory of equilibrium asset prices (Sharpe [21], Lintner [17]). On the other hand, recent models of the valuation of futures contrasts in capital market equilibrium have not considered the role of hedgers (Grauer and Litzenberger [10]).