Independent scholarship · Open accessThe monthly digest

Indexed text attachment

games-and-information-in-higher-categories.tex

Source for Games and Information in Higher Categories

View paper · Download original file

Lines 3819–3918 of 6559. Source text is displayed literally.

3819The remedies all work by letting the terms of trade depend on something correlated with type: a costly3820signal chosen by the informed side (\cref{ch:signalling}), a menu from which types select3821(\cref{ch:mechanism}), a warranty, a reputation, a certifier.3822 3823\section{Insurance}3824 3825Selection is equally adverse when the informed party is the buyer. Individuals have wealth $W$ and suffer3826a loss $d$ with probability $\pi_H$ or $\pi_L<\pi_H$ according to their privately known risk type; a3827fraction $\lambda$ are high risks. Competitive insurers offer contracts specifying a premium and a3828coverage. A contract that breaks even on the population average is a bargain for high risks and3829overpriced for low risks.3830 3831\begin{theorem}[Rothschild--Stiglitz]\label{thm:RS}\Scl[\cite{RS76}]3832In the competitive insurance market with two risk types, where an equilibrium is a set of contracts such3833that none makes losses and no further contract would be profitable if offered:3834\begin{enumerate}[label=(\roman*)]3835\item there is no pooling equilibrium;3836\item in a separating equilibrium high risks buy full insurance at their fair premium, and low risks buy3837  partial insurance at their fair premium, the coverage being the largest that high risks do not prefer to3838  their own contract;3839\item this candidate is an equilibrium if and only if the fraction of high risks is at least some3840  $\lambda^*\in(0,1)$; otherwise no equilibrium exists.3841\end{enumerate}3842\end{theorem}3843 3844The reasoning for (i) is \emph{cream-skimming}. Indifference curves of the two types cross once3845(\cref{thm:MS}): low risks are less willing to pay for additional coverage. Hence near any pooling contract3846there is another with slightly less coverage and a lower premium that attracts only low risks, and is3847profitable. For (iii): if high risks are rare, the population-average price is close to the low-risk3848price, and a pooling contract with nearly full coverage beats the separating pair for \emph{both} types;3849it would upset the separating candidate, but by (i) cannot itself be an equilibrium.3850 3851\begin{example}\label{ex:RS}3852Let $u=\ln$, $W=4$, $d=3$, $\pi_H=\tfrac12$, $\pi_L=\tfrac14$. High risks insure fully at the fair premium3853$1.5$ and have wealth $2.5$ in both states. A low-risk contract covering a fraction $\alpha$ of the loss at3854the fair premium $0.75\alpha$ leaves wealth $4-0.75\alpha$ without a loss and $1+2.25\alpha$ with one. High3855risks are indifferent between it and their own contract when $(4-0.75\alpha)(1+2.25\alpha)=2.5^2$, that is at3856$\alpha^*\approx0.29$. Low risks, who under symmetric information would be fully insured, obtain less than a3857third of full coverage: the informational externality is borne entirely by the \emph{good} risks. The3858separating pair is an equilibrium iff $\lambda\ge\lambda^*\approx0.28$ (computed in3859\cref{app:computations}).3860\end{example}3861 3862Nonexistence is uncomfortable, and it reflects the fact that the equilibrium notion here is not a Nash3863equilibrium of a well-specified game but a free-entry condition with a deviation set that is not closed3864under the reactions it provokes. Three repairs are standard. One may specify the game and allow mixed3865strategies, which restores existence \cite{DM86}. One may change the conjecture that a deviating insurer3866makes about the response of its rivals: if unprofitable contracts will be withdrawn, pooling can3867survive \cite{Wil77}. Or one may abandon the strategic story and define a competitive equilibrium in which3868every contract, traded or not, has a price that reflects the cost of those who would buy it; such3869equilibria exist generally and select the separating outcome in this model \cite{AG17}. The last is the3870modern form of Akerlof's fixed point, with the price now a function on the space of contracts rather than3871a number.3872 3873\section{Securities}3874 3875In financial markets the uninformed party is a market maker who quotes prices at which she will trade with3876whoever arrives.3877 3878\begin{example}[The bid--ask spread]\label{ex:GM}\Scl[\cite{GM85}]3879An asset is worth $0$ or $1$ with equal probability. A fraction $\mu$ of traders know the value and buy if3880it is $1$, sell if it is $0$; the rest buy or sell with probability $\tfrac12$ for unrelated reasons. A3881competitive market maker sets the ask at the expected value conditional on a buy order and the bid at the3882expected value conditional on a sale:3883\[ \text{ask}=\Prob(v=1\mid\text{buy})=\frac{1+\mu}{2},\qquad\text{bid}=\frac{1-\mu}{2}. \]3884The spread equals $\mu$. It is not a cost of doing business but a transfer: the market maker loses to the3885informed and recovers the loss from the uninformed, and the spread is the price of the adverse selection3886in the order flow.3887\end{example}3888 3889\begin{example}[Kyle's lambda]\label{ex:kyle}\Scl[\cite{Kyl85}]3890An asset has value $v\sim N(p_0,\Sigma_0)$. An insider who knows $v$ submits an order $x$; noise traders submit3891$u\sim N(0,\sigma_u^2)$; the market maker sees the total $y=x+u$ and sets the price $p=\E[v\mid y]$. Look for a3892linear equilibrium $x=\beta(v-p_0)$, $p=p_0+\lambda y$. The insider maximizes3893$\E[(v-p)x]=(v-p_0-\lambda x)x$, so $\beta=\frac{1}{2\lambda}$. The market maker's rule is a regression,3894$\lambda=\frac{\beta\Sigma_0}{\beta^2\Sigma_0+\sigma_u^2}$. Solving,3895\[ \lambda=\frac{\sqrt{\Sigma_0}}{2\sigma_u},\qquad\beta=\frac{\sigma_u}{\sqrt{\Sigma_0}},\qquad\Var(v\mid p)=\frac{\Sigma_0}{2}. \]3896The insider trades less aggressively the more her trades move the price; the price moves more the more3897informative the order flow; and in equilibrium exactly half of her information is impounded in the price,3898whatever the amount of noise.3899\end{example}3900 3901Both examples have the shape described in \cref{ch:belief}: a forward kernel from the hidden state to an3902observable (the order), and a backward map from the observable to a belief about the state (the price),3903which is its Bayesian inverse. The informed trader chooses the forward kernel taking the backward map as3904given; the market maker's backward map must be the inverse of the forward kernel actually chosen. A3905rational-expectations equilibrium is a fixed point of this pair---a Bayesian lens that is consistent with3906the best response it induces. The same structure, with a manager's unobserved action in place of the3907insider's order, is the signal-jamming model of \cref{ch:signalling}.3908 3909\section*{Notes}3910 3911The market for lemons is Akerlof's \cite{Ake70}; the lattice structure of its equilibria and their Pareto3912ranking are discussed in \cite{Wil80,MWG95}. The insurance model is due to Rothschild and Stiglitz3913\cite{RS76}; the reactive equilibrium concepts are from \cite{Wil77}, mixed-strategy existence from3914\cite{DM86}, and the price-theoretic equilibrium from Azevedo and Gottlieb \cite{AG17}. The sequential3915trade model of the bid--ask spread is Glosten and Milgrom's \cite{GM85}; the batch-auction model is Kyle's3916\cite{Kyl85}. The source text \cite[ch.~9]{Ras06} develops lemons markets through several variants with3917different distributions of quality and tastes; \cref{ex:lemons,ex:lemons2} are in that spirit with3918different numbers.