Dynamic Contracts

Prof. Lutz Hendricks

Econ720

December 1, 2011

1 / 38

Issues

Many markets work through intertemporal contracts: Labor markets, credit markets, intermediate input supplies, ... Contracts solve (or create) a number of problems:
1 2 3

Insurance: rms insure workers against low productivity shocks. Incentives: work hard to keep your job. Information revelation: you can lie once, but not over and over again.

2 / 38

Optimal contracts

If there are no frictions, agents can write complete contracts. Frictions prevent this:
1 2

Lack of commitment: borrowers can walk away with the loan. Private information: rms dont observe how hard employees work.

We study optimal contracts for these frictions.

3 / 38

An analytical trick

Dynamic contracts generally depend on the entire history of play.

"Three strikes and you are out"

The set of possible histories grows exponentially with t . A trick, due to Abreu, Pearce & Stachetti, makes this tractable. Use the promised expected future utility as a state variable. Then the current payo can (often) be written as a function of todays play and promised value.

4 / 38

Money lender model

Thomas & Worrall (1990); Kocherlakota (1996). The problem:

A set of agents suer income shocks. They borrow / lend from a "money lender". They cannot commit to repaying loans. How can a contract be written that provides some insurance?

5 / 38

Environment

The world lasts forever. There is one non-storable good. A money lender can borrow / lend from "abroad" at interest rate 1 . A set of agents receive random endowments yt . They can only trade with the money lender.

6 / 38

Preferences

t =0

t u (ct )

Note: determines time preference and interest rate.

7 / 38

Endowments

Each household receives iid draws yt . y takes on S discrete values, y s . Probabilities are s .

8 / 38

Complete markets

Households could achieve full insurance by trading Arrow securities. Consumption would be constant at the (constant) mean endowment.

9 / 38

Incomplete markets

We consider 3 frictions:
1 2 3

Households cannot commit not to walk away with a loan. Households have private information about yt . Households have private information and a storage technology.

The optimal contracts in the 3 cases are dramatically dierent.

10 / 38

Sample consumption paths

Ljunqvist & Sargent (2007)

11 / 38

Sample consumption paths

Ljunqvist & Sargent (2007) /0+)12-*, 3 !"45&), 67889:

12 / 38

How to set up the problem

Assumptions:
1 2 3

the money lender oers the contract to the household the household can accept or reject the household accepts any contract that is better than autarky

13 / 38

How to set up the problem

The optimal contract can be written as an optimization problem:

max prots subject to: participation constraints.

The state is the promised future value of the contract. To characterize, take rst-order conditions.

14 / 38

One sided commitment

Assumption: Households can walk away from their debt. As punishment, they live in autarky afterwards. The contract must be self-enforcing. Applications: Loan contracts. Labor contracts. International agreements.

15 / 38

Contract

We can study an economy with one person - there is no interaction. A contract species an allocation for each history: ht = {y0 , ..., yt } An allocation is simply household consumption: ct = ft (ht ) (1)

The money lender collects yt and pays ct .

16 / 38

Contract

Money lenders prot: P =E

t =0

t (yt ft (ht ))

(2)

Agents value: v =E These are complicated!

t =0

u (ft (ht ))

(3)

17 / 38

Participation constraint

With commitment, the lender would max P subject to the resource constraint.
What would the allocation look like?

Lack of commitment adds a participation constraint: E

t =

u (ft (ht )) u (yt ) + vAUT

walk away

(4)

stay in contract

This must hold for every history ht .

18 / 38

Autarky Value

If the agent walks, he receives vAUT = E

t =0

u (yt ) =

E u (yt ) 1

(5)

19 / 38

Recursive formulation

The contract is not recursive in the natural state variable yt . History dependence seems to destroy a recursive formulation. We are looking for a state variable xt so that we can write: ct = g (xt , yt )

xt +1 = l (xt , yt )

20 / 38

Recursive formulation

The correct state variable is the promised value of continuation in the contract: vt = Et 1 j u (ct +j )
j =0

(6)

The household enters the period with promised utility vt , then learns yt . The contract adjusts ct and vt +1 to fulll the promise vt . Proof: Abreu, Pearce, Stachetti.

21 / 38

Recursive formulation

The state variable for the lender is v . The obective is to design payos, cs and ws , for this period to max discounted prots
S

P (v ) = max s [( ys cs ) + P (ws )]
cs ,ws s =1

(7)

ws is the value of v promised if state s is realized today.

22 / 38

Constraints
1

Promise keeping:
S s =1

s [u (cs ) + ws ] v

(8)

Participation: u (cs ) + ws u ( ys ) + vAUT ; s Bounds: cs ws [cmin , cmax ] [vAUT , v ] (10) (11) (9)

Cannot promise less than autarky or more than the max endowment each period.
23 / 38

Lagrangian / Bellman equation

P (v ) = max s [( ys cs ) + P (ws )]
cs ,ws s =1 S

(12) (13) (14)

s =1 s

s [u (cs ) + ws ] v

+ s [u (cs ) + ws u ( ys ) vAUT ] Note: Participation constraints may not always bind. Then s = 0.

24 / 38

FOCs

cs ws

: s = u (cs ) [s + s ] : s P (ws ) = s + s

(15) (16)

Assumption: P is dierentiable. (Verify later) Envelope: P (v ) = What do these say in words? (17)

25 / 38

FOCs

Simplify:

u (cs ) = P (ws )1

(18)

This implicitly denes the consumption part of the contract: cs = g (ws ). Properties:
Later we see that P (v ) is concave (P < 0, P < 0). ) Therefore: u (cs ) dcs = [ P ((ws)] 2 dws and dc /dw > 0.
P ws

A form of consumption smoothing / insurance.

26 / 38

Promised value
P (ws ) = P (v ) s /s Two cases:
1

(19)

Participation constraint does not bind:

s ws

= 0 = v

Participation constraint binds:

s > 0 P (ws ) < P (v ) ws > v

27 / 38

Participation constraint does not bind

ws = v regardless of the realization ys . Consumption follows from u (cs ) = P (v )1 cs = g2 (v )

The household is fully insured against income shocks in the range where s = 0. Intuition: this happens for low y . The lender may lose in such states: he pays out the promise.

28 / 38

Participation constraint binds

The constraint: u (cs ) + ws = u ( ys ) + vAUT implies cs < y s (21) because ws v vAUT (any contract must be better than autarky otherwise the agent walks). The household gives up consumption in good times in exchange for future payos. To make this incentive compatible, the lender has to raise future payos: ws > v . (20)

29 / 38

Amnesia
When the participation constraint binds, c and w are solved by u (cs ) + ws = u ( ys ) + vAUT u (cs ) = P (ws )1 This solves for cs ws = g1 ( ys ) = l1 ( ys )

v does not matter! Intuition: The current draw ys is so good that walking into autarky pays more than v . The continuation contract must oer at least u ( ys ) + vAUT , regardless of what was promised in the past.
30 / 38

The optimal contract

Intuition: For low y the participation constraint does not bind, for high y it does. The threshold value y (v ) satises:
1 2

Consumption obeys the no-participation equation u (cs ) = P (v )1 . The participation constraint binds with ws = v : u (cs ) + v = u ( y [v ]) + vAUT

y (v ) > 0: Higher promised utility makes staying in the contract more attractive.

31 / 38

Consumption function

Ljunqvist & Sargent (2007)

32 / 38

Properties of the contract

For y y (v ): Pay constant c = g2 (v ) and keep c , v constant until the participation constraint binds. For y > y (v ): Incomplete insurance. v > v . v never decreases. c never decreases. As time goes by, the range of y s for which the household is fully insured increases. Once a household hits the top y = y S : c and v remain constant forever.

2 3 4 5

33 / 38

Intuition

With two-sided commitment, the rm would oer a constant c .

It would collect prots from lucky agents and pay to the unlucky ones. Because of risk aversion, the average c would be below the average y . The rm earns prots.

With lack of commitment:

Unlucky households are promised enough utility in the contract, so they stay. Full insurance. Lucky households have to give up some consumption to pay for future payouts in bad states. To compensate, the rm oers higher future payments every time a "prot" is collected.

34 / 38

Implications

Think about this in the context of a labor market. "Young" households are poor (low v and c ). Earnings rise with age. Earnings volatility declines with age (because the range of full insurance expands). Old workers are costly to employ. Firms would like to re them. This broadly lines up with labor market data.

35 / 38

Implications

Inequality is rst rising, then falling. Young households are all close to v0 initially. Old households are perfectly insured in the limit. Middle aged households dier in their histories and thus payos.

36 / 38

Numerical example

Outcomes as function of highest y experienced. Ljungqvist & Sargent (2007)

37 / 38

Reading

Ljungqvist & Sargent, "Recursive Macroeconomic Theory," 2nd ed. ch. 19. Abreu, D., Pearce, D., & Stacchetti, E. (1990). Toward a theory of discounted repeated games with imperfect monitoring. Econometrica: Journal of the Econometric Society, 1041-1063. Kocherlakota, N. R. (1996). Implications of ecient risk sharing without commitment. The Review of Economic Studies, 63(4), 595. Thomas, J., & Worrall, T. (1990). Income uctuation and asymmetric information: An example of a repeated principal-agent problem. Journal of Economic Theory, 51(2), 367 - 390. doi:10.1016/0022-0531(90)90023-D

38 / 38