[SOUND] We have pretty solid arguments for

why companies should pay out cash.

We just talked about a few of them, like signaling,

cash management, but we haven't talked about how we should do it.

We haven't talked about the choice between dividends or stock repurchase.

That's the topic of this lesson, okay?

Let's start with data.

What you show here or what you see here is what has happened to payout policy

from the beginning of the 1970s up to recent dates using US date.

So this is for the US.

The blue line shows you the total payout, as a fraction of net income.

You can see that this line is around from 20 to 25%, fairly stable.

Companies were paying back 20 to 25% of their cash to investors.

What has happened in the data, though, is that the red line has decreased.

The red line is the dividends, right.

In the 70s almost everything was dividends, so companies use it mostly or

almost exclusively dividends to return cash to investors.

But more recently, share purchase have picked up.

In the current dates, repurchase and dividends are equally important, okay?

So that's the summary.

It seems that now, both repurchase and dividends are equally important.

So companies are doing more and more stock repurchase and

cutting dividends, paying fewer dividends, and

keeping this 20 to 25% payout ratio fairly stable over time.

Let's think about the arguments we learned to try to think about does this make sense

in terms of taxes, right?

What we discussed is that repurchase are almost surely going to generate

fewer taxes for investors.

Investors, for one thing, they don't need to sell the stock

if they're going to face a very high taxable gain.

If they face a very high marginal tax rate, right?

With dividends there's no choice.

If you pay a dividend, all the investors are going to have to pay taxes.

So, even though both have the same statutory tax rate of 20%, what academics

researches believe is that, the stock repurchase is going to minimize taxes.

So, there might be one reason why repurchase has picked up.

Let's think about signaling.

As we discuss it, one reason to pay out, one reason

to return cash to investors is to signal that you're confident about the future.

You're confident about your ability to generate cash.

Right?

Repurchase is going to have an additional role, though.

Consider a situation when a company believes the stock price is too low,

all right?

So there is some violation of market efficiency.

The stock price is $80, but the company believes it should be $85.

What the company can do is, it can signal that to the market

by telling the market that it is willing to buy stock at $80, right?

The company goes in the market and says I'm willing to pay $80,

that is going to signal to the market that the stock price may be too low.

The company wants to buy it back, so maybe we should increase the stock price.

And notice that the difference here is that there is no such effect

with dividends.

Both spend cash.

But the dividend goes to everybody,

and there is no stock price associated with the dividend payment right?

So the dividend cannot signal that the stock price is too low.

Here's an advantage for stock repurchase.

Let me talk about another argument.

It's a reason why companies do repurchases in the real world.

This one is not a reason that I think makes sense from a welfare,

from a society point of view.

But there is evidence that it does happen, which is earnings management, okay?

Repurchases are going to allow a company to manage earnings, specifically,

repurchase allow companies to increase earnings per share.

Earnings per share are calculated as earnings divided by shares outstanding,

and what will happen for most companies in the data, what

will happen is that a stock repurchase is going to mechanically increase your

earnings per share, whereas dividends is not going to have any effect, okay.

Let me give you an example.

Suppose that the consensus analyst EPS forecast for next quarter is $3 a share.

So analysts are forecasting EPS, right?

They believe that the company's going to produce an EPS of $3 a share.

But the company's CFO realizes that your EPS is actually going to be $2.98 instead.

Maybe your profits are going to come a little bit short, right?

So you're not going to be able to meet the analyst forecast.

What should the CFO do?

One possible answer is that the CFO might repurchase shares.

This is exactly what I showed In a recent paper called the reforce and cloned.

What we've shown is that companies that expect to

fall just short of the analysts forecasts are much more likely to repurchase shares

than companies that meet the analysts forecasts without doing a repurchase.

This is what this graph shows.

Zero here, you can think of zero as the point at which

companies were going to miss the analyst forecast.

So companies that are just to the left of missing the forecast

are much more likely to repurchase shares than companies that are just to the right.

So that is unequivocal evidence,

we believe, that EPS management advise stock repurchase.

What we also find is that 37% of all stock

repurchase are conducted by companies that are just about to miss EPS targets.

We also find that companies are more likely to do this

when compensation depends on EPS targets, right?

EPS is a common way to set bonus plans in companies, for example.

And the same companies that do stock repurchase,

they end up reducing investments and deployment following these repurchase.

So our interpretation of this evidence is that it's actually a little bit troubling.

As we learned already in corporate finance in this course,

we really shouldn't be focusing so much on earnings per share.

But what happens in practice is that analysts forecast earnings per share,