Is the Value Premium Dead? - YouTube

- A value stock has a low price

relative to some fundamental metric,

like book value or earnings.

The value premium is the excess return

that value stocks are expected to earn over the market.

This excess return has historically been

and is expected to continue to be positive.

In other words, value stocks have higher

expected returns than the market.

The out-performance of value stocks

was first documented in a 1985 paper

by Barr Rosenberg, Kenneth Reid, and Ronald Lanstein

with the provocative title

"Persuasive Evidence of Market Inefficiency".

They had observed that low priced stocks

had persistently higher risk adjusted returns

than they should in an efficient market.

I'm Ben Felix. Portfolio Manager at PWL Capital.

In this episode of Common Sense Investing,

I'm going to tell you

all that you need to know

about the value premium.

(upbeat music)

Seven years after the 1985 paper

by Rosenberg, Reid and Lanstein,

Eugene Fama and Kenneth French

came out with a paper

that would forever change the study of asset pricing,

"The Cross-Section of Expected Stock Returns".

Fama and French's 1992 paper

took the empirical observations

of small cap and value stocks delivering higher returns

than their level of riskiness

relative to the market would predict,

and theorized that small cap and value stocks were exposed

to independent systematic risks,

separate from the risk of the market.

They went on to build

their now famous Fama French Three-Factor Model

for asset pricing

which includes the independent risks of the market,

small stocks and value stocks.

In 1992, Fama and French examined the 28 year data set

from July, 1963 to June, 1991 for U.S. stocks.

Over this time period,

there was a large

and statistically significant value premium,

with value stocks beating the market

by 4.9% per year on average,

that's a huge difference.

In the year 2000, Fama and French

used previously unavailable data

to examine the value premium from July, 1926 to June, 1963.

U.S. value stocks beat the U.S. market

by a smaller but still meaningful 2.39% per year

on average over this time period.

This should be compelling.

Value stocks had historically offered

an independent source of risk,

which has diversification benefits in a portfolio.

And they had offered a risk premium in excess of the market.

The U.S. value premium is well-documented

for the full period

leading up to Fama and French's 1992 paper.

But it is also well-documented

that anomalies in the financial markets

tend to disappear or decline once they've been published.

This is an important point

and it goes to the core of financial theory.

An observation like value stocks

beating the market historically,

is not helpful in isolation.

The observation becomes helpful

when there is a strong theoretical explanation for it.

The combination of empirical observations and theory

allows us to build predictions

about how things might be expected to behave in the future.

For example, if stocks that start with the letter A

outperformed the market for a given 28-year period,

there is no logical reason to believe

that the same would be true for the next 28-year period.

If, as Fama and French proposed,

value stocks have higher expected returns

because they are riskier,

We would not expect

the value premium to disappear post-publication.

Economic theory predicts that

riskier investments should demand higher expected returns.

There is also a theoretical explanation for the existence

of the value premium from behavioral economics.

Both the risk-based and behavioral explanations

for the value premium have strong empirical support.

From the risk perspective,

companies with lower prices tend to be under distress,

have high financial leverage,

and face substantial uncertainty in future earnings,

as documented by Nai-fu Chen and Fang Zhang

in their 1998 paper "Risk and Return Value Stocks".

And they tend to be much riskier than growth stocks

in bad economic times,

and only slightly less risky in good economic times,

as documented by Lu Zhang in his 2005 paper,

"The Value Premium".

These are real economic reasons

for investors to demand higher expected returns

for taking on more risk.

On the behavioral side, in their 2012 paper,

"Identifying Expectation Errors

In Value/Glamour Strategies:

A Fundamental Analysis Approach",

Joseph Piotrowski and Eric So

observed systematic pricing errors

causing value stocks to be under priced

and growth stocks to be overpriced.

For example, investors may be too optimistic

about the future growth prospects of a glamorous firm

like Apple or Tesla, irrationally bidding up their prices

and reducing their expected returns,

while neglecting value stocks,

increasing their expected returns.

Other biases like confirmation bias,

familiarity, anchoring, and loss aversion

could also explain

the persistent mispricing of value stocks.

On the surface, it might seem like a behavior-based

value premium would go away post-publication,

because investors would be expected to smarten up

once they realized their behavioral errors.

But behavioral biases do tend to be persistent,

even when we know about them.

There are also limits to arbitrage

that can make it really hard

for even the most disciplined rational investors

to take advantage of mispricings

caused by behavioral biases.

Think about the example of a pension investment officer

choosing between two funds.

One fund is expected to match the market's return

and never underperform,

while the other fund, maybe a value fund,

is expected to beat the market by 1% per year on average,

but it might have five

or even 10 year periods where it under-performs the market.

The pension investment officer likes her job

and wants to keep the pensioners happy,

so she is willing to give up

the higher expected returns of the value fund

to avoid the potential pain of under-performance.

This agency issue causes a limit to arbitrage

that will allow the mispricing of value stocks to persist.

So far, we have talked about historical data

for the value premium up to 1991,

and the theory behind why we expect to see

a positive value premium in the future.

But there's been a big problem since 1991.

The U.S. value premium has been much weaker

than it was in the past.

For the period from July, 1991 to June, 2019,

the value premium was an economically modest

1.17% per year on average,

and it was not statistically different from zero.

Fama and French came out with a new paper in January, 2020,

simply titled "The Value Premium",

where they ask, based on the weak post-publication data,

whether we can confidently conclude

that the U.S. value premium declined or even disappeared

following their 1992 paper.

They found that the high volatility

of monthly value premiums precludes us

from drawing any conclusions about a change

in the expected value premium based on this time period.

Not being able to reject the hypothesis

that out of sample expected premiums

are the same as in sample expected premiums,

sounds really nice,

but the last few years, even the last decade,

have been a bloodbath for U.S. value investors.

For the three years ending December, 2019,

value stocks trailed the market

by 6.88% per year, on average.

If that's not pain, I don't know what is.

For the decade ending December, 2019,

value stocks trailed the market

by 2.24% per year on average.

As painful as the recent history may have been

for any committed value investor,

periods like this should not be unexpected.

In their 2018 paper "Volatility Lessons",

Fama and French used Bootstrap

to simulate 100,000 time periods of varying length,

to estimate the probability

of the value premium being positive

over a given time period.

They found that over a 23 year period,

there was a 23% chance of a negative premium.

For a ten-year period,

they found a 9% chance of a negative premium.

We also have to remember

that while the value premium can trail the market

for long periods of time,

the market can similarly trail treasury bills

for a long period of time.

And here's the kicker.

Historically, there have been more ten-year periods

where the U.S. market delivered a negative premium,

than there have been ten-year periods

where U.S. value delivered a negative premium.

And guess what?

During some of those periods

where U.S. market delivered a negative premium,

U.S. value delivered a positive premium.

If that's not a diversification benefit,

I don't know what is.

This discussion is focused

on the U.S. value premium,

and whether we can still count on it

now that everyone knows that it exists.

There is strong theoretical basis.

Whether we take the risk-based or behavioral position,

to believe that a positive

expected value premium will persist.

Despite a run of relatively weak performance,

we can't conclude with any confidence

that the value premium has changed.

And even this recent period of extreme pain

falls well within what a value investor

should expect from time to time.

The recent weak performance of value

may have even created an opportunity.

I do not condone market timing or factor timing.

They are both very hard to do successfully.

Cliff Asness, the co-founder of AQR Capital

and a former student of Eugene Fama,

similarly discourages trying to time investment decisions,

most of the time.

Cliff has playfully referred to market and factor timing

as sins and suggests that investors

only sin a little and only in very specific cases.

Cliff wrote a research note in late 2019,

explaining that while factor timing

is not usually a good idea, the current value spread

makes a compelling case for sinning a little

by increasing your weight in value stocks.

The value spread is a tool that was created at AQR,

and it is used to measure the relative cheapness

of value stocks compared to history.

Using AQR's measure for value,

which includes controlling for things

like profitability and earnings quality,

as at August, 2019.

value was in the 97th percentile of historical cheapness.

And excluding the tech bubble,

it was as cheap as it has ever been by a fairly wide margin.

Unlike Cliff, I am not saying

that you should time a trade into value,

but I do think

that Cliff's point is applicable to anyone

who already has a longterm value tilt in their portfolio.

Despite the recent pain,

now might be the worst time to give up on value.

I've spent a lot of this video telling you

why we shouldn't give up on the U.S. value premium,

despite its relatively weak performance,

since it was publicized by Fama and French in 1992.

We always have to remember that while the headlines

tend to focus on the the U.S., the U.S. is not the world.

The experience of value investors in other countries

over the same time period has been very different.

From July, 1991 through June, 2019,

the same post-publication period

where we observed a declining U.S. value premium,

value stocks in Japan

beat the Japanese market by an annualized 6.14%.

Value stocks in the U.K. beat the U.K. market

by a more modest 0.29% annualized.

Emerging markets value beat the emerging markets

by an annualized 3.02%.

Asia Pacific value

beat the Asia Pacific market

by an annualized 5.4%.

Australian value beat the Australian market

by an annualized 2.95%.

And in a strange twist,

Canadian value trailed the Canadian market

by an annualized 0.53%.

Having a value tilt in a portfolio isn't for everyone.

It can be painful.

And it can be painful for long periods of time,

as we've seen.

For index investors, a value tilt also adds complexity

through the need to add additional value index funds

to a portfolio.

For anyone willing to live through a bit of pain,

and manage a bit of extra complexity,

We have seen that value stocks offer access

to a meaningfully positive expected average premium.

Recent history has been especially painful

for value investors,

but periods like this

are where the premium is ultimately earned

for those who stick with the strategy.

Thanks for watching.

My name is Ben Felix of PWL Capital,

and this is Common Sense Investing.

If you enjoyed this video,

please share it with someone

who you think could benefit from the information.

And don't forget, if you've run out

of common sense investing videos to watch,

you can tune into weekly episodes

of the "Rational Reminder" podcast

wherever you get your podcasts.

(energetic music)