[00:02:22] Rob Arnott: Sure, the genesis for the strategy was the aftermath of the dot com bubble. A dear friend of mine who ran a company called the common fund that managed commingled university endowments.

[00:02:32] Rob Arnott: He also served on the New York state pension board and several other boards. And he was just beside himself. He was very distressed because money was pouring into index funds. And index funds had a 4 percent position in Cisco at a time when Cisco was a tiny company and that subsequently went down 90%.

[00:02:54] Rob Arnott: You had the whole dot com suite of companies down. Well, NASDAQ was down 80 percent in that bear market. And so it was the crash of the dot com bubble. He came to me and he said, there’s got to be a better way. I had long thought that indexing by market cap. Cap is a great way to match the market because the market is cap weighted.

[00:03:16] Rob Arnott: But on one level a not so clever way to invest because any stock that’s overvalued relative to its future prospects is going to be overweight in the portfolio relative to a fair value weight. Any stock that’s undervalued destined to outperform is going to be underweight because you’re tying the weight to the price.

[00:03:35] Rob Arnott: Now indexers have heard that criticism since the launch of the S&P back in 1957. And they always had a ready retort to that. They said, of course we overweight the overvalued and underweight the undervalued. But, unless you can tell me, which is which, you haven’t said anything useful. Well, it turns out that acknowledging that you’re overweight the overvalued and underweight the undervalued and doing so because you’re tying the weight to the price, if the price doubles, your weight doubles.

[00:04:01] Rob Arnott: There’s a missed opportunity there. One of the originators of the Capital Asset Pricing Model, Jack Treanor, he was a dear friend, wrote a paper in the Financial Analyst Journal, Why Valuation in Different Indexes Work, and it was inspired by Fundamental Index. He pointed out that you don’t know if stocks are over or undervalued.

[00:04:22] Rob Arnott: Fair enough. But you do know that some are overvalued and some are undervalued. You do know that with cap weighting, you’re overweight the overvalued and underweight the undervalued. Take any other weighting scheme, if it ignores price, if it ignores market cap, a stock that’s undervalued might be over or under weight.

[00:04:42] Rob Arnott: And the result is that the errors cancel instead of magnifying. And he also pointed out that the price action of a stock, stocks can come into favor, and if the fundamentals don’t match the price appreciation, then watch out, there could be mean reversion. The market’s constantly changing its mind on what a company’s worth, so a stable anchor, like the size of its business, We’ll lead to a rebalancing alpha.

[00:05:08] Rob Arnott: And that’s the cool part. It does have a value tilt. Gross stocks are deemphasized down to their economic footprint. Value stocks are reemphasized up to theirs. I mean, Chevron’s a huge company, but it’s not huge market cap. Nvidia is a big company, but it’s not an enormous company and it’s priced. As if it’s truly spectacularly enormous.

[00:05:32] Rob Arnott: Current pricing is about 30 times sales. Scott McNally from Sun Microsystems back in 2002 was questioned by Congress about his stock having cratered and his comment was. We were priced at over 10 times sales. If business is steady, that means we have to deliver a hundred percent of the total revenues of the company and dividends to shareholders over the next 10 years to justify the price.

[00:06:01] Rob Arnott: And that assumes that the company has no expenses, that the shareholders are paying no taxes. He said it’s preposterous for a company to be 10 times sales. What were the investors thinking? And he was actually vocal in 99 and 2000 that his stock was way over price. Currently you have NVIDIA at 30 times sales.

[00:06:22] Rob Arnott: Is that justified? If it’s got stupendous growth over the coming decade, if 10 years from now, it’s going to be 30 times its current size, then maybe. But, absent that, it does make sense to weight it lower. So, with Fundamental Index, an idea I’d been playing with for, in the back of my mind, for at least a decade, was, why not weight companies by their sales?

[00:06:48] Rob Arnott: Why not weight them by their book value? And so we tested a whole array of methods. Measures, sales, profits, cash flow, EBITDA, book value, dividends, number of employees, where McDonald’s would be one of your largest holdings. It didn’t matter which measure you used. We found it added about 2 percent per annum.

[00:07:12] Rob Arnott: It did introduce a value tilt, because if you’re weighting companies on their fundamental size, you’re de-emphasizing growth, you’re re-emphasizing value. Now these are better companies, for sure. But it’s in the price. The market’s already priced in the quality of the business, and unless it exceeds lofty expectations, it’s not going to help you.

[00:07:33] Rob Arnott: The value stocks. Troubled companies with headwinds. It’s not going to hurt you unless it underperforms bleak expectations. So it does have a value tilt, and the result is that when you compare it with the cap weighted market, fundamental index wins over long periods of time by about 2 percent a year. If you compare it with value indexes that are cap weighted, you get the same drag from the cap weighted version, and so relative to value indexes, we find that the outperformance is relentless.

[00:08:05] Rob Arnott: In the U. S. in the last 17 years, RAFI would have beat the Russell Value Index 14 out of 17 times. The global index against MSCI ACUI value outperforms in 15 out of 17 years. I mean, that’s astonishing consistency. And it all comes about not because of the value tilt. The value tilt is a consequence of fundamental waiting and yes, value does have a little bit of an alpha, but the big alpha comes from rebalancing, from concentrating against the markets, constantly changing opinions.

[00:08:40] Clay Finck: Yeah, and since you mentioned Nvidia, I just looked up what their waiting was in the S&P 500 and it’s 6.25%. So anyone that’s buying that index today is putting a 6.25% of that say a hundred dollars into.

[00:08:54] Clay Finck: And I think most people would assume that a strategy like this has underperformed, at least in recent years, because this sort of strategy would have been underweight many of the biggest companies in the S&P 500. So, how would you explain what is compensated for being underweight the bigger winners?

[00:09:13] Rob Arnott: If you look back, value peaked in 2007 and it hit bottom in the summer of 2020. It bottom bounced again at the end of 21. It’s bottom bouncing again now. Back in 2007, growth stocks were three times as expensive as value. Summer of 2020, it had widened to nine times. That means value had gotten three fold cheaper in the space of those 13 years relative to growth stocks. That spread of 9 to 1 was actually wider than at the top of the dot com bubble.

[00:09:46] Rob Arnott: Now value underperformed, if you look at Russell value versus the market, it underperformed by 38 percent. But if you’re getting cheaper by two thirds and you’re underperforming by 38 percent, it means that if the relative valuation had remained where it was, Value would have outperformed. So, in fact, the value effect is alive and well.

[00:10:05] Rob Arnott: I wrote a paper in the Financial Analyst Journal in 2021, Reports of Values Death Have Been Greatly Exaggerated, in which I looked at the attribution and found that the entire drawdown for value came not because value companies were struggling, Not because the value investing was a bad idea, but because value investing was falling out of favor and getting cheaper.

[00:10:28] Rob Arnott: When value is melting down to that extent, no, RAFI will not beat the broad market. But it beat value handily. And the result is from 2007 to today, RAFI has roughly kept pace with the broad market. Similar return to the S&P 500, similar return to the Russell 1000, but anchoring on value stocks, which are now direct cheap.

[00:10:52] Rob Arnott: So to the extent anyone believes value deserves a place in their portfolio, why would you invest in value any other way than fundamental index? The relative performance versus value is relentless. The relative performance against the broad market is more iffy. When value’s winning, boy, do we win handily, big time.

[00:11:15] Rob Arnott: When value’s losing, we have headwinds and we often will underperform. The aftermath of the global financial crisis, value staged. A nice comeback in the middle two quarters of 2009, and then started to sputter. Value, therefore, roughly matched the stock market in the year 2009, underperforming big in the first quarter, then outperforming, then underperforming, finishing right about in line with the market.

[00:11:45] Rob Arnott: How did RAFI do? As value was tanking, we went to a deeper value tilt. And when value snapped back, we got it back with a vengeance. We had a 2 percent position in B of A, 2 percent position in Citi. Why? Because those companies were about 2 percent of the U. S. economy each. We had a 1 percent position in GM because it was 1 percent of the U S economy.

[00:12:11] Rob Arnott: It went bust about eight weeks later. And so the stock went to zero, but if one stock goes to zero and two other stocks, triple you’re doing fine. So the shocking thing was that RAFI. lagged the market in 2008 when value was getting crushed by 3%, and then outperformed in 2009 by 15%. So you got it back with a vengeance.

[00:12:36] Rob Arnott: And we see that happen again and again. Anytime I see RAFI underperform, I start to get excited because I know what happens next. We pivot into a deeper value tilt, and when value comes back, we get it back with a vengeance.

[00:12:49] Clay Finck: Now to my understanding, RAFI refers to this fundamental approach, and we have many different ways this could be applied, so there’s many different RAFI strategies, so maybe you could discuss what some of those strategies look like, how they’re sort of constructed, and then how the positions are weighted in these RAFI strategies.

[00:13:08] Rob Arnott: The purpose of RAFI is to be an economy weighted index that weights companies according to their macroeconomic footprint. Now, when we walk on the beach, our footprint in the sand has multiple measures. Length, width, depth of the footprint. Same thing is true for companies. Chevron has a fairly large footprint in terms of sales relative to all publicly traded companies.

[00:13:32] Rob Arnott: Fairly large on profits relative to all company profits, very large on dividends relative to all company dividends, and pretty large on book value. NVIDIA has a big economic footprint on profits, but not on sales. And so if there’s any mean reversion to their monumental 53 percent profit margin, Watch out.

[00:13:56] Rob Arnott: Book value. They’re small. Dividends. Well, they don’t pay any. So you wind up with a company with good size economic footprint on one measure and a small footprint on three others. Pretty good size company. Certainly in the top 25, but not in the top three. And Chevron reliably in the top 10. So you wind up with weightings that reflect the macro economy.

[00:14:19] Rob Arnott: Now, the way I like to think about this is if you wait the portfolio, if you choose stocks based on how big they are, the companies, and weight them on how big they are, you have a fundamental index. RAFI stands for Research Affiliates Fundamental Index. The cap weighted market will take all the growth stocks with lofty expectations and say, these are great companies, I’m pricing them higher.

[00:14:45] Rob Arnott: So relative to the macro economy, the cap weighted market is a growth tilted, momentum chasing, Popularity weighted index, which works great when you have momentum, when you have people chasing the current fads and trends, and it’ll struggle. So I like to think of fundamental index as an economy weighted index.

[00:15:11] Rob Arnott: You weight companies on how big their business is. If the price rises and the fundamentals don’t, you rebalance down. If the price tanks and the fundamentals stay solid, You rebalance up. So you’re rebalancing against price moves that aren’t supported by changes in the fundamentals. Now, the market is cap weighted, so the cap weighted indexes studiously mirror the look and composition of the market.

[00:15:39] Rob Arnott: Fundamental index studiously mirrors the look and composition of the macroeconomy. Now, think of it this way. From the vantage point of cap weighted indexers, This is not a passive strategy. This is a, an active value tilted strategy that contra trades against price moves. Fair enough. Viewed from the perspective of fundamental index the cap weighted market is hardly passive.

[00:16:03] Rob Arnott: Cap weighted market is making constant changes in What it believes the future will look like, and where it believes the opportunities will lie. And so, the cap weighted market, from a fundamental perspective, is a growth tilted, momentum chasing, popularity weighted index, which is going to work when trends are chasing popular ideas and can hurt you badly. If those trends turn out to be a bobble.

[00:16:31] Clay Finck: So you put together this write up on the equity risk premium and nine myths related to that. So I wanted to touch on this article, but I think a good place to start here would be to talk about what is the equity risk premium and how do you think about calculating this metric?

[00:16:49] Rob Arnott: Sure and just in case your audience is curious, 50th anniversary edition of the journal of portfolio management. The founding editor was a dear friend of mine. A large chunk of the paper was focused on the equity risk premium. People look at past differences in returns, stock returns versus bond returns, or stock returns versus cash returns.

[00:17:14] Rob Arnott: I going to say that’s the risk premium. No, that’s the past historical excess return. That’s a different beast entirely. The past historical excess return has a lot of constituent parts. Dividend yields versus bond or cash yields can be very different. The growth rate of earnings and dividends. Can be significant, and the growth rate of coupons on bonds, zero by definition.

[00:17:42] Rob Arnott: So, very different constituent parts to the returns, and a big part of the return can be revaluation. In the spring of 2009, the U. S. stock market hit bottom at 13 times the 10 year average earnings for the companies in the S&P or Russell Index. That’s called a Shiller P. E. ratio of 13. Today it’s three times that, it’s 38 times.

[00:18:09] Rob Arnott: So the stupendous run up over the last 15 years is in large measure a consequence of rising valuation multiples of people being willing to pay more and more and more for every dollar of earnings, that’s not risk premium. That’s excess return. Risk premium is what your expected return is for stocks relative to bonds or cash.

[00:18:33] Rob Arnott: And if you want to be objective about it, you’d want to calculate today’s risk premium by looking at differences. You’ve got a yield of close to 5 percent for cash. You’ve got a yield of one and a quarter percent for stocks, second lowest in history, only the dot com bubble had a lower dividend yield.

[00:18:53] Rob Arnott: And we know what happened after that. Alright, well that’s a big difference, but surely, you’re going to see wonderful growth in earnings and dividends. Well, the consensus according to Lipper is 18 percent long term growth in earnings and dividends. Pardon me, we’re at a peak in earnings and dividends as a percentage of GDP, very near historic all-time highs.

[00:19:16] Rob Arnott: And you’re expecting 18 percent growth? Are you expecting 18 percent GDP growth? I don’t think so. If you’re expecting 5 percent nominal, 2. 5 percent real GDP growth, then 5 percent nominal growth in earnings or dividends ought to be a reasonable expectation, not 18. Now you aren’t going to get any growth from an income from bonds or cash.

[00:19:41] Rob Arnott: They’re called fixed income for a reason. So, all right, you’ve got one and a quarter yield. You’ve got 5 percent as a reasonable growth expectation. That gives you a six and a quarter percent return. Compare that with, call it four and three quarters or four and a half for U. S. cash. And you’re looking at about a one and a half percent risk premium for stocks.

[00:20:03] Rob Arnott: That’s the risk premium. Now, people can have subjective views. That’s also a risk premium, but it’s subjective and it’s based on whims and speculations and often some pretty shabby thought. So, there is an equity risk premium right now, but it’s kind of skinny. So, there’s another component. I mentioned earlier that the rebound in valuation multiples since 2009 Was stupendous Shiller PE ratio tripled.

[00:20:31] Rob Arnott: Dividend yields fell by two thirds. Alright, well that’s a big change in what people will pay for every dollar of earnings or every dollar of dividends. People are willing to pay a lot. And so then the question is, where will that relative valuation be? Let’s say 10 years from now. We have a website.

[00:20:50] Rob Arnott: People look up Asset Allocation Interactive. It’s a free website where we forecast returns for upwards of 200 different asset classes. And our forecast for U. S. stocks is about three and a half percent. I just took you through arithmetic that said six and a quarter. What’s the difference? Well, let’s assume today’s Shiller P.E. ratio is fair, and 10 years from now it’s still 38 times. Then you get your six and a quarter percent. Give or take. Because growth will be a little different from the five, but not drastically. Let’s suppose you get mean reversion, the historic norm is 18 to 20. And we think there’s a new normal, that is to say, markets are sustainably more expensive than they used to be 30, 40, 50 years ago.

[00:21:34] Rob Arnott: So we think the normal is about low 20s, 23, let’s say. If it falls from 38 to 23, that’s going to cost you about 6 percent a year. Ouch. So let’s split the difference. Let’s say maybe it doesn’t mean revert or maybe it does. That takes away three and gets you to about three and a half. So that’s the way we do our calculations of the risk premium.

[00:21:55] Rob Arnott: And on that basis, the equity risk premium today is modestly negative. How can you have a negative risk premium? Another part of that paper speculated on the idea that It’s not a risk premium. It’s a fear premium. It was mislabeled all along, which misled the quant community and the academic community for the last 60 years.

[00:22:17] Rob Arnott: It’s a fear premium and viewed from a fear perspective. Back in 2009, people were afraid of equities because they’d just fallen by half. Today, people are afraid of cash because cash is Been a useless asset. And so from today’s perspective, the fear premium ought to be negative.

[00:22:38] Clay Finck: Yeah. I mean, one of the big takeaways for me from that article is to be mindful of this bias people have, where they look at what happened the past 10 years and assume that that’s what the next 10 years are going to look like. But these cycles play out into where, you know, after one stupendously good decade of ever expanding multiples, you should be ready for some multiple contraction after that.

[00:23:00] Rob Arnott: Were the 90s followed by a stupendous decade? No, they were followed by a lost decade. Was the last decade followed by another last decade?

[00:23:09] Rob Arnott: No, it was followed by a stupendous decade. And so when you have a bull market, that’s taken us to valuation multiples that have only been seen once before in history, top of the. com bubble, you really ought to ratchet down your return expectations. I’m not saying. Be massively pessimistic and shun stocks forever.

[00:23:29] Rob Arnott: No, but do you really want to bet that this market’s going to double when it could just as easily or perhaps more easily have a meaningful beer market? Personally, I’d like to have some margin of safety.

[00:23:43] Clay Finck: So one of the myths you highlighted in your article is that long term forecasting is hard. How can it be that short term forecasts, for example, one year, are largely unpredictable, but somehow longer term forecasts, which are really just many one year’s strung together all into one, those are actually predictable, how does that work?

[00:24:04] Rob Arnott: Long horizon forecasts, the long term return of anything you invest in, consists of what’s the yield, what’s the growth in income in the years ahead, And what’s the change in valuation multiples, or in the case of bonds, what’s the change in the spreads? And if you know those three numbers, you can estimate future returns with remarkable precision.

[00:24:28] Rob Arnott: For instance, 10 year returns for 10 year bonds, your best forecast is the starting yield. You’re going to be within 1 percent either way. For the stock market, it’s not as formulaic, but the starting yield plus long term historical growth plus some mean reversion towards the past gets you a forecast that shockingly over the last 50 years would have been Within two or two and a half percent per annum of the actual stock market return most of the time.

[00:25:00] Rob Arnott: Well, that’s pretty impressive. Now, the same dynamics play out in the coming year. The coming year, you’ve got yield, you know that. You’ve got growth with a lot more uncertainty, a lot more wiggle room. So yeah, the growth could be five, but it could be 10 or it could be zero. So big uncertainty. And the revaluation component.

[00:25:23] Rob Arnott: Could be up 30 percent or down 30 percent relative to earnings. That’s why the one year forecasts are very difficult. There’s also the interesting issue that using long term history can lead you astray. My favorite example is the second half of the 20th century. Now, a half century for any investor is a long term.

[00:25:45] Rob Arnott: Most of us died before our first 50 years of investing are wrapped up. And so, 50 years is a long time. The stock market Went from a dividend yield of 8 at the start of 1950 to a dividend yield of 1. 1 at the end of 1999. Half a century, 7 fold change in the value attached to every dollar of dividends.

[00:26:10] Rob Arnott: Sevenfold change over the course of 50 years, that’s 4 percent per annum. So that means that for a half century, your historical return had a biased picture of returns. Pensions use a pension return assumption. in determining what the pension impact on earnings per share is. And that pension return assumption can be somewhat subjective.

[00:26:37] Rob Arnott: At the beginning of 2000, the average pension return assumption was nine and a half percent. The highest in history, was never higher before, never higher after, and it was at a peak. So people were using past returns to say, Oh, I can make it nine and a half percent. I have a balanced portfolio, my bonds are yielding 6, my stocks are going to give me 12, so I can assume 9.5 for my 6040 portfolio. And lo and behold, stocks with a dividend yield of 1, to earn that 12 percent return assumption would have had to see growth of 11 percent a year and would have had to see no mean reversion in valuations. So of course the earnings didn’t grow 11 percent a year, and of course there was mean reversion.

[00:27:22] Rob Arnott: So, long term history, even long term history, can lead you astray. I’ve been stunned that in what’s now a long career, going back 47 years, I’ve been pounding the drum on re-evaluation for a long time, for decades. As something that misleads us, it’s misleading us big time now because you have a tripling evaluation multiples in the last 15 years.

[00:27:49] Clay Finck: So it’s as you’ve alluded to earlier in the conversation, growth stocks have led the charge and really led to this big revaluation in the market overall. So when looking at, say, growth and value and where that disparity sits, like, where are we at today in the cycle and how does that compare to past cycles?

[00:28:10] Rob Arnott: The short answer to your question is nobody knows where we are in that cycle. So I’m going to guess, and it is, I’ll acknowledge, only a guess. My guess is that we’re very near the bottom of a value cycle and value today just returning to historic norms. Thanks. Where value is normally about a fourth to a fifth as expensive as growth would require value to double relative to growth stocks, double 10, 000 basis points of incremental performance.

[00:28:42] Rob Arnott: Spread that over 10 years, and you have 20 plus percent outperformance per year. Do I think we’re going to see 20 percent outperformance per year over the next 10 years? No, I shaved that a bit. But I do think value could easily beat growth by 5, 6, 7 percent a year for the next 5 years, and therefore beat the market by 2 or 3 percent a year for the next 5 years.

[00:29:06] Rob Arnott: That’s a big win. I think that’s a conservative forecast. Do I want to forecast that value will beat growth in 2025? Sure, but not with high confidence. I’d give it 60, 40 odds on a next year basis, 40 percent chance of being dead wrong. And I give it at least 80, 20 odds in the coming 10 years.

[00:29:27] Clay Finck: And when looking at how these cycles turn, does the market need some sort of catalyst to spark this reversal or do you need some sort of recession or how do these cycles turn?

[00:29:39] Rob Arnott: They often turn with a catalyst, but not always. I’ve asked the question, what was the catalyst that burst the dot com bubble, other than gravity? Things were too high. I’ve heard a lot of answers, but none that were totally satisfying as something that could cause A complete crash of growth stocks. So I think it was overwhelmingly just gravity.

[00:30:03] Rob Arnott: But usually there’s a catalyst and the funny thing about catalysts is that by definition, they’re a surprise. If it’s not a surprise, it’s not a catalyst. So it’s kind of a fun parlor game to ask the question, what could be the catalyst? Because one catalyst could be. Mean reversion in valuations, which is another way of saying gravity.

[00:30:23] Rob Arnott: Another catalyst could be slow down in the economy in which growth stocks mean revert towards value stocks and relative valuation. Another catalyst could be very similar to the dot com bubble and burst. In the dot com bubble, the narrative was these companies are building a new future. They’re creating the internet, and the internet’s going to be huge.

[00:30:48] Rob Arnott: It’ll change how we buy and sell goods and services. It’ll change how we communicate with our clients and with our friends. It’ll change how we socially interact. It’ll change how we get our news. All true. Narratives have the advantage of being mostly true. What wasn’t true was the narrative went on to say this is going to happen astonishingly fast.

[00:31:09] Rob Arnott: And these companies are going to be the leaders ten years from now, too. And disruptors get disrupted. Palm was briefly worth more than General Motors in 2000. Does anyone have a Palm Pilot anymore? Does anyone have anything that vaguely resembles a Palm Pilot or is made by the company anymore? Of course not.

[00:31:27] Rob Arnott: It was disrupted by BlackBerry. BlackBerry was already on the scene and BlackBerry went on to have dominant market share in handheld devices by 2008 and nine, and then the iPhone, which came along in 2007, disrupted BlackBerry. So you had sequential disruptions. Google disrupted other search engines. Ask Jeeves, remember Ask Jeeves?

[00:31:51] Rob Arnott: So the whole notion that these companies have a lock on the future is dangerous. Disruptors get disrupted. The whole notion that the world will embrace AI fast. is a little dangerous. AI is going to be huge. But it’s going to take time, because people embrace change slowly. Luddites will want to slow it down.

[00:32:14] Rob Arnott: But it’s very nature will be to happen fast, but human nature is to embrace change slowly. So I’m guessing, just like the internet, 2005 was very similar to 2000 in the way we all behaved. 2024 is very different and the same thing will happen with AI. I think five years from now, it’ll be making noticeable strides but it’s not going to be an unrecognizable world. 2050 might be an unrecognizable world to those of us here today.

[00:32:43] Clay Finck: When I look at through much of the 2010s and the later 2010s, many people would justify higher stock prices simply by just looking at bond yields and how low bond yields were. So it’s like, hey, if the S&P 500’s paying a 2 or 2.

[00:32:57] Clay Finck: 5 percent yield and the bond yields One or 2%, then, you know, I’ll take the S&P 500 all day. It was how a lot of people justified that. And what’s sort of taken me by surprise is just how strong the market’s been after interest rates have come up. So that justification just sort of goes out the window. So it seems to me that in your view, what’s led a lot of this market rally is just this narrative around AI. Is that really how you characterize it?

[00:33:25] Rob Arnott: There’s that. There’s also the political effect. A lot of people think that reducing the burden of regulatory overreach may be in the cards. The other political element that’s a fun one.

[00:33:39] Rob Arnott: I wrote a paper eight weeks before the election in which I showed that with contentious elections, the stock market rallies from a few days before the election for the next four to six weeks. No matter who wins, it doesn’t matter who wins. I had a billionaire client in 2004, very recognizable name, who was a big donor to Democrats.

[00:34:03] Rob Arnott: And he came to me just a few weeks before the election. And he said, I’m getting scared. I think GW could win and I want to pull out. I want to get out of the stock market. And. My comment to him was there’s people on both sides who feel that way. There are some who think, Oh my god, Kerry might win. I want to get out of the market.

[00:34:26] Rob Arnott: And when the election happens, one side or the other is relieved, and the ones who were so alarmed that they got out of the market are coming back in. So, my suggestion is, wait, no matter who wins, I predict the market will go up, and six, eight weeks after the election, let’s have another conversation. If you want, Al, Let’s take you out.

[00:34:48] Rob Arnott: But the bottom line is that’s a fairly persistent pattern. And it’s kind of a fun one, no matter where you sit politically, somebody on the other side of the spectrum shares your sense of alarm about what’s coming. I mean, every election is portrayed as the most consequential of our lifetimes. Every election that’s ever happened, the country’s survived just fine over the next four years. So rather than being alarmed, look for opportunities. It’s kind of fun.

[00:35:14] Clay Finck: I wanted to get to next, but I had one more comment I wanted to make here because I thought it was just so interesting. I was tuning into one of your previous conversations and you were discussing NVIDIA, which we’ve already touched on today.

[00:35:27] Clay Finck: And one of your comments that I loved regarding NVIDIA was, I don’t have a problem with 70 times earnings, I have a problem with 40 times sales. 40 times sales is very hard to justify, which I thought was quite interesting because so many investors are so anchored to, you know, what is the company earning today?

[00:35:44] Clay Finck: What’s that earnings going to look like into the future? I believe you also said the Achilles heel for NVIDIA is going to be compressed margins, not revenue growth. They’re going to be a great business going forward. It’s just like, how much margin are they going to be able to get? I was curious if you could talk more about that.

[00:35:58] Rob Arnott: The narrative about NVIDIA is this company is changing the world, and it is. This company dominates the super chips market that in turn dominates AI, 100 billion annual sales, 60 billion annual profits. That’s phenomenal. But the narrative is AI is going to be a big part of our future. The sales are going to keep soaring.

[00:36:24] Rob Arnott: Absolutely true. If you model it out, and if you assume, let’s say 30 percent growth in sales per annum compounded for the next decade, are the competitors going to sit on their hands? I don’t think so. Is the price going to stay 40, 000 per chip? I don’t think so. So let’s speculate on a future where the growth, the unit growth is 30 percent a year.

[00:36:46] Rob Arnott: Yeah. But instead of 90 percent market share, they got 40 percent in 10 years. Instead of 53 percent profit margin like they had last quarter, they’re down to 30 percent profit margin. Plug those numbers in and what you find is that the earnings in 2034 are almost identical to the earnings in 2024. Now, that’s It’s shocking to people who haven’t thought about it in those terms.

[00:37:12] Rob Arnott: The simple fact is, often bubbles are blown based not on a flawed narrative, but on narrative overreach. The competition is not sitting on their hands. AMD already has chips that are as fast as NVIDIA’s fastest. Why don’t people switch? Because the price is one third as much. They don’t switch because you got to reprogram everything.

[00:37:36] Rob Arnott: And so, switching is not easy. Switching can happen over the course of 2, 3, 5 years. And so, I look on NVIDIA as a beautiful company with a great vision, and Wang was an absolute genius in saying these superchips that we’re about to build Let’s leave some dead real estate on them for future innovations.

[00:37:58] Rob Arnott: And his team pushed back and said, what are you talking about? That’s a waste. And he said, no, we’re going to need it. And sure enough, the innovations filled in that space and they were. Instantly the dominant player in superchips. Well, that’s very cool. And AI is very cool. Anyone who’s played around with the AI tools, AI’s been around forever.

[00:38:18] Rob Arnott: I was doing neural nets in the 1980s. But they weren’t very useful back then because the compute power wasn’t large and the data, you need massive data. The new phenomenon is user friendly AI. And the AI that’s at our disposal now is very powerful. Average person doesn’t have to be a nerd. You can just pose a question to ChatGPT.

[00:38:42] Rob Arnott: You can say, I want some graphic arts to Dali, and describe roughly what you want, and it’ll spit out ten images that are interesting. It’s going to change our world, but the leaders of today won’t necessarily be the leaders of tomorrow, and they will have competition. There will be priced impression.

[00:39:01] Clay Finck: Yeah. It’s also quite humbling to look back at 1999 and see what many of the big companies were then who the market darlings were and how those stocks favored over the next 10, 20 years. So I wanted to be sure to get to your write up on next the upside of getting dumped. So it’s often said that a benefit of index investing is that by owning the index, you’re automatically adding to the winners and cutting your losers.

[00:39:26] Clay Finck: And Being the contrarian that you are, you’re interested in even looking at the losing stocks that get removed and are ignored by the market and turning away from the winning stocks that replace the companies that get dropped out of the index and actually tend to not favor as well. So please walk us through your research on next.

[00:39:45] Rob Arnott: Sure, well, firstly, the title is obviously a whimsical one. The upside of getting dumped. I’m reasonably sure you’ve never experienced getting dumped, but I have, and most of us have. If you get dumped, you can wallow in self-pity for life, or you could pull your socks up and move forward and learn lessons and seek to do better next time.

[00:40:03] Rob Arnott: Okay, companies can do the same thing. If a company gets dumped from the S&P, it’s a troubled company. It’s out of favor. It’s unloved. It’s cheap. And I like to joke, it may go on to achieve great failure. Or it may get its act together, and it’s not priced for that. It’s priced for continued struggles and perhaps oblivion.

[00:40:27] Rob Arnott: We created the index next, in end of June. And the goal is to find stocks dropped from the largest 500, very similar to the S&P. And the largest thousand, very similar to the Russell 1000. We did research on those two indexes, and we found that the stocks dropped by those indexes, on average, outperformed by 28 percent, 2, 800 basis points over the next five years.

[00:40:54] Rob Arnott: That’s a big margin of victory. It doesn’t matter whether you use S&P or Russell or a filter of top 500 and top 1000, The mechanism for dropping a stock is the same. It’s out of favor. It’s plunged in valuation multiples. It’s plunged out of your target market cap range and is replaced with a frothy highflyer.

[00:41:15] Rob Arnott: And so we launched the index in end of June. There’s a stock Luma that was dropped by the S&P and I believe April of 2023 dropped by the Russell index in June of 2023. And signed a major contract with Meta about 8, 10, 12 weeks ago to help Meta beef up their AI capabilities. And so the stock is up 550 percent since June.

[00:41:47] Rob Arnott: Well, that’s pretty cool. We hold stocks for 5 years. Why? Because it’s a persistent trend. It slows during the 5 years. I mean, it starts out 7 percent a year and then it’s 2 or 3 percent a year. And so we give it up after five years. Stocks deleted over the last five years from the top 500 or top 1, 000, there’s 150 of them, and we equally weight them.

[00:42:09] Rob Arnott: Basically, we’re playing a game not unlike fundamental index. These companies tumble in market cap, market value, but they’re still not small companies. And the result is, fundamental index chooses companies based on how big the business is and then weights them on how big the business is. Cap weighting doesn’t matter the size of the business.

[00:42:30] Rob Arnott: If it’s out of favor and unloved and dirt cheap, it’s gone. So, when we look at the 150 names, Lumen was one of them, and that single stock has given us over 200 base points of alpha because it’s quadrupled as a share of the index in just a few weeks. That’s cool. That’s fun. And you don’t need a lot of examples like that out of 150 names and you don’t need a lot of examples like that out of 150 names.

[00:42:55] Rob Arnott: It’s not the only one that’s more than doubled. And so you just need a handful of companies that regain their footing to beat small cap value, which is the appropriate benchmark for next. ETF architects is a company that launches ETFs and they launched an ETF based on the next index back in early September.

[00:43:16] Rob Arnott: And it’s still small, but it’s very powerful way to access a part of the market that has no natural owners. Index funds have dumped it. Index funds, when they sell, they have to sell to active managers. Active managers don’t like these companies. That’s why they’re cheap. So they have to sell to companies that, active managers who don’t want it.

[00:43:39] Rob Arnott: And so it exits the index abnormally cheap. And that’s part of what sets the stage for the subsequent tendency for outperformance. It wins historically, testing it back over the last 30 years, wins about half of the time, 15 out of 30 years, give or take. But when it loses, it loses by about 5%. And when it wins, it wins by about 18%. That’s an asymmetry that I love.

[00:44:01] Clay Finck: I want to ask you about a few numbers if you happen to know them. So first, I’m curious of a company that’s in the S&P 500, what amount of the shares are held by the index? And then my second question is, when a company gets added to the S&P 500, what sort of boost does it get to the multiple?

[00:44:20] Rob Arnott: We’ve documented this again and again. I did a paper back in 1986. Entitled: S&P Additions and Deletions, A Market Anomaly. And my understanding is that S&P actually used that paper, or S&P indexers used that paper to lobby S&P to preannounce additions and deletions, which they started doing in October of 1989.

[00:44:45] Rob Arnott: In that paper I showed that additions outperform in the immediate aftermath of being added by 5 percent or more, and deletions underperform by 5 percent or more. Might not have been 5 each way, I think the total gap was about 8. The gap by 2017 was now 16%. So additions win, and deletions lose by about 16%, relative to one another.

[00:45:12] Rob Arnott: So that stretching of the rubber band snaps back in the aftermath. And that’s a beautiful thing. So, the 2017 paper, Buy High and Sell Low with Index Funds, Pointed out, firstly, that when you buy in an index fund, typically you’re going to be buying a frothy company that’s expensive and your buying pressure will push it higher.

[00:45:35] Rob Arnott: If you sell, get rid of a company, typically your selling pressure will push it lower. And that gap is about 16%. Now, indexes do most of their trading in a single block trade on the change date, the effective date. Why? Because they’ve trained the world to see tracking error relative to the index as a sign of incompetence.

[00:46:00] Rob Arnott: So even if the tracking error is positive, shame on you, it could work negatively next time. So you’re clearly not competent. So they are much more interested in locking in the exact price at which it’s added or dropped from the index. Back in 1986, they couldn’t do that. It was Not preannounced. It was announced typically on a Wednesday right after the market closed and the effective price was the one that had just closed.

[00:46:25] Rob Arnott: All right, so that meant index funds moved the prices and therefore underperformed the indexes that they were tracking and lobbied aggressively, come on guys, preannounce this so that we don’t have to underperform. And These days, some of them say, see, we don’t move prices. No, they do not. They themselves, but the hedge funds that front run those trades and then flip the stocks to the index funds, they’re leaving money on the table, quite a bit of money.

[00:46:55] Rob Arnott: So the process of adding and dropping companies is a big deal. It’s getting bigger and bigger because index funds are getting bigger and bigger. I don’t have the number for your first question, how big are indexes as a share of market cap. I think it’s in the ballpark for S&P 500, I think they own somewhere in the ballpark at 25 percent of the total market cap of every stock in the S&P 500, and of course zero of every stock that’s not.

[00:47:21] Rob Arnott: So if a stock is added to the index, they have to go from zero to own 25 percent of the total market cap of the company. And if it’s dropped, they have to go from owning 25 percent to owning zero. A lot of them will want to do that trade in a market on close block trade on the effective date to lock in exact matching of the index.

[00:47:46] Rob Arnott: And they don’t care if hedge funds have pushed the price up on the addition or pushed the price down on the deletion because they’re trying to lock in zero tracking error. Well, that’s too bad. The 2017 paper, we posed a really simple question. What happens if S&P makes a change? You write it on a post it note.

[00:48:05] Rob Arnott: You stick it on your fridge. And a year later, you do the trade. You wind up with 20 basis points higher returns than the S&P. Because you don’t participate in that huge melt up in price and melt down in price on the deletions. Well, Nixed captures one side of that, and Rafi captures the other. Both sides of that, because you own companies, you’re going to add a company long after a cap weighted index adds it, because a cap weighted index will add it when it’s frothy, expensive, popular, beloved, but still a small company.

[00:48:39] Rob Arnott: And RAFI will add it when it’s proven its merit and is now a big company. Deletions. Stocks get kicked out of the cap weighted index before they’re small companies, when they’re small cap, and get kicked out of RAFI after they’re no longer big companies. So, the turnover associated with additions and deletions is lower for RAFI than it is for cap weight.

[00:49:04] Rob Arnott: There is a rebalancing component that adds back into the turnover, but one of the beauties of all of this work is all interconnected. There are vulnerabilities in the way cap weighting works, and that’s not to say cap weighting is a bad idea. It’s totally fine.

[00:49:19] Clay Finck: Many people are probably tired of me or us discussing the S&P 500, but it’s just so popular for people to talk about.

[00:49:26] Clay Finck: I had a question related to the S&P 500 is that many people see this as. Just a passive approach to investing in the market, but at the end of the day, somebody has to select what companies are versus aren’t included in the S&P 500. I was curious if you could just talk about what that process typically looks like for an index like that.

[00:49:48] Rob Arnott: For the Russell 1000, the process is formulaic. If it’s, I think if it’s gets into the top 900 by float or out of the top 1100 by float, it’s added or dropped. Something like that. Or S&P, it’s a committee decision. Tesla’s a beautiful example. March of 2020, people started to speculate that Tesla, which had just finished its third profitable quarter in a row, it had been unprofitable quarter by quarter, year by year, since inception.

[00:50:20] Rob Arnott: And suddenly it was profitable three quarters in a row, and they knew that S&P wouldn’t add it without four profitable quarters in a row. Thank you. And so people in March began to say, wow, this is about to be added and it’s big. So people started buying it. Fourth quarter profits is reported. S&P is kind of paralyzed.

[00:50:41] Rob Arnott: They aren’t sure what they want to do. And fifth quarter comes along. Pressure’s intense. They’ve got to do something. And so they decided to add it in one fell swoop in December of 2020. Well, between the decision day, which was late November and the addition day, which was mid-December, the stock was up well over 40%.

[00:51:03] Rob Arnott: But roll the clock back to March. It was up something like 250 percent all on the back of not fundamentals. Shocking to the upside, although they were. Showing an ability to turn a profit, which is cool, but based on the narrative that this stock is going to get added and there’s going to be a 25 percent of its total market cap is going to have to change hands in one fell swoop.

[00:51:25] Rob Arnott: So, I’m not critical of S&P’s methodology. I’m not critical of Russell’s methodology. I would say that the cap weighted indexes claim to be passive. People think of them as passive. They are passive. When a stock’s in the index, It stays there, and its weight in the index moves up and down with price. And the only thing that’ll change that is if there’s new shares issued or a stock buyback.

[00:51:52] Rob Arnott: But basically, it’s passive, except at the bottom of the list where stocks are added and dropped. And there it’s very active with massive trading costs. And thank God the turnover is so low because those massive trading costs are multiplied by three to five percent turnover which is tiny. So, it costs tens of basis points, and you can recapture that.

[00:52:16] Rob Arnott: You can recapture that by trading late. You can recapture that by constructing the index in a different way. So, indexes are a wonderful thing in terms of allowing you to be a free rider. Let the market do price discovery for you. Let the market decide where to allocate capital, and you just piggyback along and pay almost nothing.

[00:52:38] Rob Arnott: But, on the other hand, it’s not without its vulnerabilities, and it does leave a rebalancing alpha on the table, which Rafi captures. And it does add frothy stocks and drop on love stocks, which Nixed captures. So these are all kind of interconnected.

[00:52:54] Clay Finck: It’s fun stuff. Well, Rob, this is a fun chat. I really appreciate you coming on the show and uncovering just these interesting strategies with the Fundamental Index and NXT.

[00:53:05] Clay Finck: I want to give you a chance to give a handoff to the audience if they’d like to learn more about you and research affiliates and everything you guys are up to.

[00:53:12] Rob Arnott: Well, firstly, if you go to our website, There’s a ton of information we’ve written. I’m going to scare your audience off. We’ve written 400 journal articles in the last 25 years.

[00:53:22] Rob Arnott: And a lot of that is readily accessible. We have asset allocation interactive, which is an interactive tool that helps you look at multiple asset classes around the world and ask where’s the opportunity. Spoiler alert. The opportunity is not so much in us large cap growth stocks or in mainstream us bonds and cash.

[00:53:45] Rob Arnott: Those are priced to give you call it three to 5 percent returns. If you want good returns, it’s an opportunity rich environment because. There are markets that are priced really pretty cheap all over the world. They’re just not U. S. stocks and bonds. So U. S. value looks pretty good. Small cap looks pretty good.

[00:54:04] Rob Arnott: Small cap value looks very good. And international stocks and emerging market stocks look pretty good. Aren’t I aware that international and emerging markets have fallen far behind the U.S.? Yeah, that’s why they’re cheap.

[00:54:17] Clay Finck: Alrighty, Rob. Thanks again. I really appreciate the opportunity and hope we can do it again sometime in the future.

[00:54:23] Rob Arnott: This was fun. Thanks for the invitation.

[00:54:25] Outro: Thank you for listening to TIP. Make sure to follow We Study Billionaires on your favorite podcast app, and never miss out on episodes. To access our show notes, transcripts, or courses, go to theinvestorspodcast.com. This show is for entertainment purposes only. Before making any decision, consult a professional. This show is copyrighted by The Investor’s Podcast Network. Written permission must be granted before syndication or rebroadcasting.