Myers Stewart C 2000 Outside Equity Journal Of Finance 55 1005 1037

The main questions in the paper are:
- Why do corporations voluntarily pay dividends when the threat of takeover is remote?
- How can outside investor fund capital investments when managers can intercept (future) cash flows?

Considering positive-NPV ventures, investments demand:
- intangible assets (provided by insiders): ideas and human capital.
- tangible assets (provided by outsiders or outsiders/insiders): cash.

However outsiders may walk away with their resources or deny their use to insiders. So insiders may lose such resources in the next period ( = reduced cash flows).

In order to keep outsiders’ investment, insiders should pay a dividend (in each period) JUST SUFFICIENT to retain outsiders’ participation for at least one more period.

In seems natural that insiders do not want neither to pay less than “sufficient” (they would lose outsiders’ participation) nor to pay more than “sufficient” (because insiders would have reduced resources available).

So the question is: How to determine the “sufficient” dividend?

Myers presents two models to find it:
- “Partnership Model”: Insiders and outsiders make a deal. Outsider investors receive “now” a dividend and commit not to withdraw their assets until the next period. It is like a partnership agreement that limits outsiders’ property right (of walking away).

- “Corporation Model”: Firms will survive if today’s dividend generates a rational expectation of sufficient future dividends. Outsiders’ property right are not constrained. The control is by voting or takeover.

Another question is raised: Why should insiders improve assets’ value if they (insiders) have no control over these assets in the future?
One way is to work out this problem is to reduce outsiders’ bargaining power. How?
Issuing shares => dispersion of ownership => the exercise of control becomes costly.

(The author does not consider the choice between debt and equity)

It is assumed that cash flows are not verifiable, so contracts to prevent insiders from capturing cash flows would not work. However, location and use of assets are assumed verifiable.

Firm is a startup venture.
Insiders provide human capital or intangible assets.
Outsiders provide most of the money to buy operating assets.
Outside equity is protected only by the “primitive right of ownership”.
Outsiders can walk away with their share of assets but cannot prevent insiders from taking operating cash flows.

The capital is denoted by K.
K generates a perpetual cash flow C = mrK (r is the opportunity cost of capital and m>1 is the value added by insiders).
So, the firm’s NPV contributed by insiders is (m-1)K. This means that if insiders walk away the firm’s assets will worth only K.
Insiders have I dollars and must raise K - I from outside investors.

Insiders will capture part (or all) of the future cash flows.
Zt represents the amount captured at date t.
Yt = C – Zt is the residual cash flow paid out as dividend.
x is outsiders’ fractional ownership.
So, outsiders get xYt and insider equity gets (1-x)Yt.
If outsiders leave the firm at this moment, they take out asset worth xK.

Vtex is the ex-dividend PV of shares if outsiders do not walk away. Such investors will wait only if:
x(Yt + Vtex) >= xK

If the firm continues, the same amount will be paid out each period. Thus:
Vtex = Y/r

And we have: x(Y +Y/r) = x(Y(1+r)/r) >= xK

Insiders minimize Y to maximize Z, so:
Y = rk/(1+r)
Vtex = Y/r = K/(1+K)

The solution is not affected by x (proportion of outsiders’ participation).
A greater ownership share of insiders has no effect on dividends or on Z (the part of the cash flow captured by insiders). This conclusion is different from that one in Jensen and Meckling (1976), who state that the higher the fraction of outsiders’ shares the more insiders take private benefits.

If outsiders leave they get xK. So, insiders pay out just enough that the cum-dividend value of outsiders’ shares is xK. The dividend is set as: xK =x(Yt+ Vtex)
being that Vtex depends on the next period’s cum-dividend market value which will be equal to K regardless of permanence of outsiders. The present value is K(1+K) at t. Substituting this value for Vtex in x(Yt + Vtex) >= xK, we get Yt = K – K /(1+r) = rK/(1+r).

Insiders try to maximize NPV by raising K – I from outsiders and investing the full amount K. The insider’s net present value at startup is:
NPV (to insiders) = -I + PV(C) – xK/(1+r)

Outsiders will demand a fractional ownership x = (1+r)(K-I)/K. That is, insiders must put in enough to absorb the discount to the market value of outsider’s shares. The minimum is I – rK/(1+r).

Substituting for x,
NPV = -I + mK – (K-I) = mK-K
Which is maximized when m=1 at the margin for the last dollar invested.


There are not partnership agreements but outsiders can vote out management at any time.
This model introduces two main changes: costs of collective action such as bargaining among outsiders and insiders and the fact that the firm can continue only if the value to outsiders from doing nothing is at all times no less than the net payoff ot throwing the managers out.

Net value reachable by outsiders is alfa*K, where alfa < 1 represents the cost of organizing to vote to replace management. Thus, the ex-dividend value of outside equity cannot fall below alfa*K at any time if outsiders are to be kept from taking control. That is, expected future dividends must have a PV of at least alfa*K.

The assumptions in the model are:
- Bargaining is replaced by sequential actions. Insiders pay dividend; outsiders pocket it and decide whether it was good enough to justify their permanence for one more period.
- Payoff to outsiders from taking over is x*alfa*K, alfa<1 (Recall: it was xK in the Partnership Model).
- Outsiders have majority voting control.
- The firm has an indefinite life.
- (As in Partnership Model), firm’s assets are observable and any attempt by insiders to sell them and depart with the proceeds is verifiable and can be stopped.

- Some comments about Dividends:
The corporation can continue under current management only if outsiders believe that they receive dividends of r*alfa*K at each future date t.
PV of future dividends is r*alfa*K/r = alfa*K, so outsiders are always as well off continuing with current management as taking over.
Outsiders will not accept less than alfa*K because they can take control and realize alfa*K. Thus, if they expect a dividend r*alfa*K at t+1, then the ex-dividend value of their shares at t must be
Vtex= (r*alfa*K + alfa*K)/(1+r) = alfa*K

Outsiders will continue if the expected future dividend is greater than r*alfa*K, i.e:
Et(Yt+1) = r*alfa*K

The key to the equilibrium is the link between the current dividend and the expected future dividend. Myers (2000) assumes that Et(Yt+1) = Yt; then if the firm pays Yt < r*alfa*K, outsiders will take over.

- Will insiders bail out?
To assess the possibility of insiders bail out, the author considers the situation in which the insiders hold (1-x)% of outstanding shares.
In this case, they (insiders) receive C – Y + (1-x)Y = C – xY per period if they continue
On the other hand, they receive (1-x)K = their share of the assets, if they stop

The payoffs are:
Stop: C + [(m -1)K]/(1+r) + (1-x)K
Continue: C – xY + (m-x*alfa)K

(m-x*alfa) is the PV of insiders’ future cash flows (including their share of future dividends).

Insiders continue if m >= (1-x) + x[alfa*(1+r)2 – 1]/r.
This equation means that insiders may stop inefficiently if outsiders’ bargaining power is too high. If alfa = 1, insiders continue only if m > x(1+r) + 1, i.e. if they can earn more than the cost of capital. If insiders add no value (m=1), they stop unless alfa < 1/(1+r).
If managers become inefficient (m < 1), the payoff to stopping is C + (1-x)K. But they will continue if m > (1-x) + x*alfa*(1+r).

Insiders contribute the NPV of (m-1)K but they get more than this. However, there is no inefficiency; all NPV projects are undertaken if outside financing is feasible and insiders can coinvest.

The author also analyzes the relationship between insider and outsider investors when the firm invests in R&D.
The probability of research success is p.
If the project is worthwhile, then:

NPV = -K + pmk/(1+r)2 + (1-p)K/(1+r) >=0

The private investor can sell out and take the firm public at t = 1. If it does, however, some value is transferred to insiders because exercising control is more costly for public outside investors: they can only reach alfa*m*K of the ex post value, where alfa < 1. The private equity investor can get all the value (alfa = 1).

Jensen and Meckling (1976) stated that increased equity ownership by insiders reduces agency costs. This assumes laws or contracts that go beyond the property rights assumed by Myers (2000).

Myers (2000) points out that fraction of shares held by insiders has no effect on corporate resources taken by insiders. This is due to dividend payout policy. Insiders have to adjust payout when their actions affect Vtex(the continuation value of equity). Therefore insiders have to pay in every period to maintain their position. The amount they have to pay is given by an intertemporal constraint. Given cash flow and payout, insiders’ take is fully determined. There is no reason to spend money to monitor. Jensen and Meckling (1976) basically do not consider dividend payout policy.

Myers (2000) assumes that outsiders have complete rights to assets. So, outsiders do not need to monitor the amount or disposition of cash flow. They only need to watch the dividend paid relative to asset value.

In terms of monitoring and agency, the paper reachs the following conclusions:
Monitoring cannot be done by insiders, and dispersed outside investors cannot monitor efficiently on their own.
Since the act of investment is frequently nonverifiable, monitoring is necessarily imperfect, but also indispensable if the firm’s growth opportunities are valuable. But monitoring cannot be designed or operated to protect assets-in-place only. It has to watch how cash flows are used. In many cases the purposes of expenditure will not be obvious: apparent investments may actually benefit insiders, and operating costs may actually be investments.
If outside equity rests on property rights to the firm’s assets, then the difficulty of verifying reinvestment of operating cash flow leads to monitoring of investment. Thus, monitoring leads to agency costs, not the other way around. Once monitoring systems are set up, they inevitably lead insiders to try to capture private benefits rather than cash. This leads to the well-known agency problems emphasized by Jensen and Meckling (1976).

The main differences between “Partnership” and “Corporation” Models are:
- Corporation Model requires an infinite or indefinite horizon; Partnership Model does not.
- In Corporation Model, some positive-NPV investments cannot be financed if outsiders’ bargaining power is too high. In Partnership Model, all positive-NPV investments are undertaken since insiders have some ability to coinvest.

Finally, it is worthy noting that:
- Outside equity depends on an intertemporal constraint: the dividend paid today must ensure outside investors’ participation for at least one more period.
- Outside equity works only if insiders contribute part of the financial capital, although the contribution may come as sweat equity – the willingness to work for less than an opportunity wage.
- Firms whose future value depends on human effort and risk taking may go public to reduce the bargaining power of outside equity investors.
- It can be much more difficult to confirm new investment than the existence of assets in place. Therefore, outside equity investors must monitor the disposition of operating cash flows among operating costs, investment, and rewards to insiders. Monitoring then leads to agency costs, not vice versa.

Unless otherwise stated, the content of this page is licensed under Creative Commons Attribution-ShareAlike 3.0 License