"The Nifty Fifty" - yesterday and today

[KinAlberta]

I have credit whoiswarrenfor posting this 1998 AAII Jeremy Siegel, Nifty Fifty Revisited article (see reference below).

 

Current thoughts from the participants of this board would be welcome.

 

 

 

From the article:

 

"In 1975 there was no way of knowing which explanation was correct. But 25 years later we can determine whether the Nifty Fifty stocks were overvalued in 1972. Examination of their subsequent returns shows that the second explanation, roundly rejected by Wall Street for years, is much closer to the truth. A portfolio of Nifty Fifty stocks purchased at the peak would have nearly matched the S&P 500 over the next 26 years. ... "

 

http://www.aaii.com/journal/article/valuing-growth-stocks-revisiting-the-nifty-fifty

 

 

 

As an aside, I've always wondered if Warren Buffett hadn't long ago figured out that an index of stocks would outperform all other indexes if he could just weed out the companies that cause the churn and slowly accumulate (at opportunistic prices) just own the most permanent of growth companies. 

 

A question I once asked Burton Malkiel was basically, how could an arbitrary number of 500 stocks beat every fund manager? Could not someone sitting and holding a subset or larger set, say the S&P 278 or S&P 402 or S&P 555, etc. beat the S&P 500?

 

 

I've decided my favourite ever deal would be if these guys bought Lancashire ;)

 

 

 

In fact, I don't want to see anyone buy Lancashire, as I think its worth is far higher than what anyone out there would be willing to pay.  It's a quality company and I think the market systematically undervalues such entities (and overvalues crappy ones).  I think this explains why investors didn't necessarily overpay for the quality companies that made up the "Nifty Fifty", even though they looked very expensive on static valuation metrics (averaging c.42x P/E in 1972, more than twice the then P/E of the S&P500).  Yes, their stock prices subsequently fell, but over the subsequent years their fundamentals shone through so that a basket of Nifty Fifty stocks bought at the peak would have performed as least as well as the market to date (I suspect performance has been much better). See this article, written in 1998 http://www.aaii.com/journal/article/valuing-growth-stocks-revisiting-the-nifty-fifty

 

As for Leucadia, as opportunistic value investors they will seek to capitalise on market dislocations or misfortunes begetting individual situations.  I severely hope Lancashire doesn't fall into either of these categories (!!) and that both companies can continue to grow their intrinsic values over time.

 

Yes

[oddballstocks]

I have had some people tell me that there is no bad price paid for a good company.  I guess at some point if you wait long enough maybe it'll work out?

 

The question I have is "what is a good company?" or "what is great?"

 

Maybe a better conclusion is that market leaders rarely completely fail, and given a long enough time frame an investor will do alright with a market leader.  The question is whether investors have the conviction to hold regardless of price paid.

[thepupil]

I love this article and have read it a few times over the years. One problem with it is that many of those same companies were pretty richly valued at the time of the article's publishing; the valuations o megacap growthy companies in the late 90's helped justify the rich valuations of the Nifty Fifty.

 

For example, KO was $38 per share in August 1998 (when the article was published). It's $41 now and performed quite poorly on an absolute basis if you purchased (or did not sell it) then.

 

I don't have the time or desire to go through the rest, but I would imagine you would find similar examples in other companies. There needs to be a "re-visiting of the re-visiting of the Nifty Fifty".

 

I'm all for a long term approach and time arbitrage. But I think the article's premise is a little flawed.

[KinAlberta]

Years ago Bill Gates commented on the high valuation of Microsoft (at the time) saying something to the effect that that it was crazy because Microsoft had nowhere near the potential longevity of Berkshire Hathaway... And that basically the market didn't appreciate BRK.

[KinAlberta]

I love this article and have read it a few times over the years. One problem with it is that many of those same companies were pretty richly valued at the time of the article's publishing; the valuations o megacap growthy companies in the late 90's helped justify the rich valuations of the Nifty Fifty.

 

For example, KO was $38 per share in August 1998 (when the article was published). It's $41 now and performed quite poorly on an absolute basis if you purchased (or did not sell it) then.

 

I don't have the time or desire to go through the rest, but I would imagine you would find similar examples in other companies. There needs to be a "re-visiting of the re-visiting of the Nifty Fifty".

 

I'm all for a long term approach and time arbitrage. But I think the article's premise is a little flawed.

 

Ahh but with the dividends and the lack of tax on capital gains for a buy and hold investor, etc... I wonder how they have performed on an aftertax basis by today.

[Pelagic]

I recently read Howard Mark's book The Most Important Thing and he too made mention of the "Nifty Fifty" and the expectations associated with them at the time. He did concede that the Nifty Fifty actually outperformed the S&P 500 (I don't recall the time period he used) almost entirely thanks to the growth of Phillip Morris. The lesson of the Nifty-Fifty as I see it isn't so much the relative overvaluation associated with some of the stocks at the time, it's that a couple truly great stocks can make up for a lot of mediocre ones, even if they were all overvalued when they were purchased. Now, how do we find the next Phillip-Morris in today's "Nifty-Fifty"  ;)

[Guest Schwab711]

I recently read Howard Mark's book The Most Important Thing and he too made mention of the "Nifty Fifty" and the expectations associated with them at the time. He did concede that the Nifty Fifty actually outperformed the S&P 500 (I don't recall the time period he used) almost entirely thanks to the growth of Phillip Morris. The lesson of the Nifty-Fifty as I see it isn't so much the relative overvaluation associated with some of the stocks at the time, it's that a couple truly great stocks can make up for a lot of mediocre ones, even if they were all overvalued when they were purchased. Now, how do we find the next Phillip-Morris in today's "Nifty-Fifty"  ;)

 

Takeaway I've always had is that if you can identify the PM's then you don't have to worry about valuation most of the time. 20x - 25x is actually cheap/reasonable if the company can grow earnings 15% over 15 years or something. Obviously identifying this is tough but it seems like the lowest risk way to earn 15%+ returns over long periods. It also seems possible.

[Uccmal]

I have had some people tell me that there is no bad price paid for a good company.  I guess at some point if you wait long enough maybe it'll work out?

 

The question I have is "what is a good company?" or "what is great?"

 

Maybe a better conclusion is that market leaders rarely completely fail, and given a long enough time frame an investor will do alright with a market leader.  The question is whether investors have the conviction to hold regardless of price paid.

 

It would be more valuable to determine if anyone actually held onto these for 25 years.  My guess is not many.  Could we (they) have held onto Philip Morris through all the lawsuits, mergers etc. 

[jschembs]

Thanks for posting.

 

What I find most worrying is the degradation in standards for what's considered a quality growth company. Back then, the Nifty Fifty "had proven growth records, continual increases in dividends (virtually none had cut its dividend since World War II), and high market capitalization."

 

Compare that to many of the "buy and forget" great companies today (Visa, Mastercard, big pharma, Google, Apple, the "roll ups"). Few offer much in the way of dividend yield, preferring to utilize share repurchases to return capital. For a company trading at 30+ earnings, I think I'd much prefer them to boost dividends. Given much of these buybacks are fueled with debt, it'd be interesting to see EV/EBITDA or EV/FCF for today's buy and forget companies compared to the Nifty Fifty.

 

Further, the more speculative buy and forget companies (Amazon, Salesforce, Netflix, Zillow, Workday) haven't established any track record of generating superior (or even value-creating) returns on invested capital, yet they are certainly given the benefit of the doubt by Mr. Market. I'm sure there were equally speculative names beyond the Nifty Fifty more comparable to these names, but the broader theme in my mind is that what equity investors are willing to accept has largely continuously degraded over the years. For example, stock dividend yields in the first half of the 20th century routinely were in excess of corporate bond yields. Now certainly part of this is reasonable - better financial reporting, more oversight and disclosures, a general increase in comfort among the populous in owning future earnings streams as opposed to "gentlemen prefer bonds."

 

However, as corporate finance theory continues to eschew the notion of free cash flow (which I'm certainly a proponent of), I think we're getting further away from what the goal of that free cash flow is - to return cash to investors. Buybacks at any price is not a reasonable approach in my opinion, but that's a different discussion. Perhaps it's part of a more worrying development - demographics and the difficulty to grow an increasingly large economy makes growth in general very difficult to achieve, so investors are hoodwinked into "we need to plow earnings back into the company to grow," even if the realistic long-term growth rates may be unachievable.

 

Anyway, just a tangent I've been thinking about lately.

[ItsAValueTrap]

However, as corporate finance theory continues to eschew the notion of free cash flow (which I'm certainly a proponent of), I think we're getting further away from what the goal of that free cash flow is - to return cash to investors.

 

Buybacks versus dividends:  It's not a big difference either way, buy buybacks are arguably more tax-efficient.

 

I don't think that Visa buying back shares at a P/E of 30 is a terrible idea.  Neither brilliant nor stupid.

 

Because interest rates are really low right now, the debt-fueled buybacks kind of make sense.

 

Free cash flow: With netflix, valuing the business on current free cash flow doesn't make that much sense.  It's the nature of their business that they have lose money in the beginning, just like the cable channels did when they were starting out.  I don't think FCF is the right way of valuing cable channels and online subscription services.

 

Salesforce has plenty of free cash flow.  If you look at cash from ops - capex and divide that by the share count, you see that number grow over time. 

 

2- Maybe I'm crazy but I think a portfolio of V, MA, NFLX, Z, AAPL, big pharma, AMZN, CRM, WDAY will do alright.  The level of fraud is really low and the valuations are reasonable.  The quality of those companies is above average.

[Patmo]

I recently read Howard Mark's book The Most Important Thing and he too made mention of the "Nifty Fifty" and the expectations associated with them at the time. He did concede that the Nifty Fifty actually outperformed the S&P 500 (I don't recall the time period he used) almost entirely thanks to the growth of Phillip Morris. The lesson of the Nifty-Fifty as I see it isn't so much the relative overvaluation associated with some of the stocks at the time, it's that a couple truly great stocks can make up for a lot of mediocre ones, even if they were all overvalued when they were purchased. Now, how do we find the next Phillip-Morris in today's "Nifty-Fifty"  ;)

 

Takeaway I've always had is that if you can identify the PM's then you don't have to worry about valuation most of the time. 20x - 25x is actually cheap/reasonable if the company can grow earnings 15% over 15 years or something. Obviously identifying this is tough but it seems like the lowest risk way to earn 15%+ returns over long periods. It also seems possible.

I'd argue that it's not the lowest risk way if the identification part is tough. You can get similar returns by thoughtlessly buying a basket of purely cheap securities based on the metric of your choice, and it is a pretty easy task to identify them. You know better than I do the number of studies that have shown that. Why jump over higher hurdles?

[Jurgis]

You can get similar returns by thoughtlessly buying a basket of purely cheap securities based on the metric of your choice, and it is a pretty easy task to identify them.

 

I am not sure I agree with this.

 

"similar returns" - what returns?

"it is a pretty easy task to identify them" - is it? Are you suggesting a value-weighted index fund? I have not seen people have great returns with "thoughtlessly buying a basket of purely cheap securities" despite any studies that indicate otherwise.

[Pelagic]

 

I'd argue that it's not the lowest risk way if the identification part is tough. You can get similar returns by thoughtlessly buying a basket of purely cheap securities based on the metric of your choice, and it is a pretty easy task to identify them. You know better than I do the number of studies that have shown that. Why jump over higher hurdles?

 

We're kind of back to the question brought up earlier in the thread though. If you find a basket of say 50 purely cheap securities why not go with a smaller sub-set based on a metric like cheapness or quality etc. If you can identify a basket of 50, why not just pick the 18 best or the 12 best and go from there, this doesn't seem like adding much of a hurdle once you come up with the criteria to select the initial 50. Personally, I'm torn on this and in practice tend to gravitate toward a more concentrated approach although I'm not sure if it's correct.

 

For one, the selected basket method has some appeal, especially if your primary metric is "cheapness". While an imperfect analogy, think of it like the way venture capital funds invest, a lot of losers, a couple breakevens or small gains and a home run, if they're lucky. If you buy a basket of the absolute cheapest companies you can find, hopefully some revert to something close to what you estimated fair value to be but the appeal of a larger basket is that maybe one or two will not just reach fair value but really take off.

 

Maybe you're right, keep the hurdle low and relax. But, the temptation of identifying great compounders is always there and perhaps more than anything its an intellectual pursuit and that's why so many investors put in the time and effort to try and find great companies.

[Patmo]

You can get similar returns by thoughtlessly buying a basket of purely cheap securities based on the metric of your choice, and it is a pretty easy task to identify them.

 

I am not sure I agree with this.

 

"similar returns" - what returns?

"it is a pretty easy task to identify them" - is it? Are you suggesting a value-weighted index fund? I have not seen people have great returns with "thoughtlessly buying a basket of purely cheap securities" despite any studies that indicate otherwise.

1.Similar returns to what was suggested by the poster I answered to, not sure why you have to ask...

2.Sure, that's one way. But I am talking conceptually. What is harder, identifying a company that has a low EV/EBITDA, or a company that will compound at 15% for 15 years? The question answers itself.

3.Your personal observation is great, but it's not because nobody does it that it doesn't work. Whether you agree or not doesn't matter, it's a well documented phenomenon.

[Jurgis]

If you think that you can get 15% annual returns by mechanically buying a basket of low EV/EBITDA stocks, have fun, good luck.

 

If this worked, this forum would be empty.

 

 

[Patmo]

If you think that you can get 15% annual returns by mechanically buying a basket of low EV/EBITDA stocks, have fun, good luck.

 

If this worked, this forum would be empty.

 

If it didn't, this forum and the term "value investing" wouldn't exist.

[Jurgis]

If you think that you can get 15% annual returns by mechanically buying a basket of low EV/EBITDA stocks, have fun, good luck.

 

If this worked, this forum would be empty.

 

If it didn't, this forum and the term "value investing" wouldn't exist.

 

IMHO, this forum is not about mechanically buying anything.

 

However, if you seriously have a mechanical strategy that gives 15% annual returns by mechanically buying a basket of low EV/EBITDA stocks, yes, people would use it. Can you share it with us?

[Patmo]

If you think that you can get 15% annual returns by mechanically buying a basket of low EV/EBITDA stocks, have fun, good luck.

 

If this worked, this forum would be empty.

 

If it didn't, this forum and the term "value investing" wouldn't exist.

 

IMHO, this forum is not about mechanically buying anything.

 

However, if you seriously have a mechanical strategy that gives 15% annual returns by mechanically buying a basket of low EV/EBITDA stocks, yes, people would use it. Can you share it with us?

 

The pool of cheap securities outperforms which is why value investors look at this pool to build their books in the first place. If value didn't work, there would be no value investing. You can figure out whichever way you want to rationalize not looking at this specific area of the market for your own means, as they say there's more than one way to skin a cat, but I'm merely stating the facts. You can argue that the earth is flat all you want, it simply isn't.

[thepupil]

I think Jurgis is taking issue with your return number, not whether or not value outperforms over time.

 

15% over time is reee - donk - you - lous and should not be expected of any mechanical process.

 

 

[Jurgis]

I think Jurgis is taking issue with your return number, not whether or not value outperforms over time.

 

15% over time is reee - donk - you - lous and should not be expected of any mechanical process.

 

Exactly.

[KinAlberta]

I think Jurgis is taking issue with your return number, not whether or not value outperforms over time.

 

15% over time is reee - donk - you - lous and should not be expected of any mechanical process.

 

Exactly.

 

Hmm. In business, there's little that can't be quantified. Though much isn't broadly quantified ( like market penetration by product line, etc.) So why couldn't a mechanistic process perform at least at the 15% level.  It's just a lot of data would have to be manually collected and added to a database.

 

I also imagine some non-standard time metrics would need to be used to seek expected returns beyond more typical investor time horizons.

 

Think about how minimalist most indices are. A fixed number of securities, often just capitalization weighted/based, etc. Even fundamental indexing is pretty simplistic in design and "value" indexes are based on some pretty old views differentiating value from growth. Yet guys like Buffett and Miller have said that value and growth are two sides of the same coin.  Small and midcap indices abandon successful investments because of their size. Pulling the flowers and keeping the weeds, as Peter Lynch once said, seems irrational to me.  So I figure there has to be a more creative indexing to substantially beat a broad market.

[AzCactus]

To me, mechanically earning 15% seems pretty tough.  I would have a hard time thinking of many mutual funds that have averaged 15% for any prolonged period of time.

 

 

[Cigarbutt]

Nifty-Fifty re-visited.

http://www.fortunefinancialadvisors.com/blog/price-is-what-you-pay-value-is-what-you-get-nifty-fifty-edition

The angle taken by the author has flaws and limitations but potentially useful conclusion.

Long term active thinking is not what it used to be.

 

"The lesson from this exercise, I believe, is that investors should always be conscious of starting valuation when placing their bets.  With few exceptions, eventually valuations that are simply too high will drift back down to more reasonable levels, often at the expense of poor intermediate-term performance.  This appears to be true no matter how revolutionary the new business appears to be, and no matter how much potential you believe it has.  Of course, if your conviction is such that you plan on holding your shares for multiple decades, valuation may indeed matter less to long-term returns, but that is assuming you follow through on your commitment.  Over several years of sub par performance, that is much easier said than done."

 

A lesson I got too is that the best long term performers have been selling carcinogenic and coronary artery blocking products.

 

[cameronfen]

Nifty-Fifty re-visited.

http://www.fortunefinancialadvisors.com/blog/price-is-what-you-pay-value-is-what-you-get-nifty-fifty-edition

The angle taken by the author has flaws and limitations but potentially useful conclusion.

Long term active thinking is not what it used to be.

 

"The lesson from this exercise, I believe, is that investors should always be conscious of starting valuation when placing their bets.  With few exceptions, eventually valuations that are simply too high will drift back down to more reasonable levels, often at the expense of poor intermediate-term performance.  This appears to be true no matter how revolutionary the new business appears to be, and no matter how much potential you believe it has.  Of course, if your conviction is such that you plan on holding your shares for multiple decades, valuation may indeed matter less to long-term returns, but that is assuming you follow through on your commitment.  Over several years of sub par performance, that is much easier said than done."

 

A lesson I got too is that the best long term performers have been selling carcinogenic and coronary artery blocking products.

 

I think "highly addictive" had more to do with the long term performance. 

[Cigarbutt]

I think "highly addictive" had more to do with the long term performance.

 

Agreed. Probably the best scenario involves selling addictive products that kill slowly or pills that stabilize chronic diseases without curing them.  ::)

Did you know that Coca-Cola used to contain cocaine? The ingredient has been substituted with happiness. :)