In this episode of The Acquirers Podcast, Tobias chats with Vivek Viswanathan, portfolio manager of the Rayliant Quantamental China Fund. During the interview Vivek provided some great insights into:
- Value Performs Well In China
- Why Chinese Companies Don’t Like To Report Losses
- Stripping Out Value Components To Generate Better Returns
- Quantamental Investing
- Momentum Does Not Perform Well In China
- China’s SOE Reforms
- Factors That Do Well, Continue To Do Well In China
- How China Cut-Off Short Selling
- Investing In China “A” Shares
- Foreign Holdings Released Daily In China
- Chinese Regulators Require A Lot Of Disclosure
- Trade Sizes Released Daily In China
- A-Shares vs H-Shares
You can find out more about Tobias’ podcast here – The Acquirers Podcast. You can also listen to the podcast on your favorite podcast platforms here:
Tobias: Hi, I’m Tobias Carlisle. This is The Acquirers Podcast. My special guest today is Vivek Viswanathan, PhD. He is the portfolio manager of the Rayliant Quantamental China Fund. The ticker is RAYC. We’re going to be talking about the peculiarities of investing in China A-shares right after this.[The Acquirers Podcast theme]
Tobias: Tell me a little bit about Rayliant.
Vivek: Absolutely. In 2016, Jason Hsu, and I, and a few others spun off the Asia arm of Research Affiliates to start Rayliant Global Advisors. Rayliant is primarily focused on China A shares, and emerging markets as well as alternative investments within China including commodity futures, bonds, and futures. We’ve launched a China A ETF, active ETF, RAYC, in December of last year. We’ve been managing a mix of Chinese equities and alternatives in China.
Investing In China “A” Shares
Tobias: Let me just ask you, let’s just take a step back. What is a China A-share?
Vivek: You can access Chinese equities through a variety of mechanisms. One is onshore China. That’s where you go through Hong Kong connect, and you buy shares on Shanghai or Shenzhen. These shares tend to be much more inefficiently priced. They’re traded primarily by retail investors. It’s about 80% retail trading in China A-shares. There are other ways to access Chinese equities. Some firms list in Hong Kong.
These are called either H-shares, red chips, or P chips, depending on their SOE status and where they’re incorporated. You also have China ADRs listed in the United States. They each have very different behavior. ADRs, and H-shares. red chips, and P chips tend to be a little bit more efficient. They’re traded by fairly sophisticated investors. Hong Kong investors and US investors tend to be relatively sophisticated. A-shares are exceedingly inefficient, which is nice.
If you look at US large cap stocks for example, 90% of active managers underperform. If you look at EM, about 55% of managers underperform. It could be as high as 65%. In China A-shares, despite having a 1.5% average management fee, the onshore funds tend to outperform. They tend to outperform the large cap index by about 3% and just barely edge out the full market as a whole. But that’s a good space to be. It’s nice to be in a space where it is– You’re among folks who are outperforming. You have a counterparty to the trade that is less rational and those tend to be the retail investors. There’s a larger wellspring of alpha there.
Chinese Regulators Require A Lot Of Disclosure
Vivek: There’s also another beautiful aspect of China which is, they have reams of data. So, regulators require a lot of disclosures. They require the exchanges to seek out data on firms and understand more about their accounting, so on and so forth, and they will even create reports that can be publicly read about given firms. This coupled with the fact that there are so many retail investors who are not reading any of this data.
They’re not accessing any of this data. It makes it excellent for institutional investors. Indeed, we see that manifest, not just in mutual funds, but in foreign investors. Foreign investors in China A-shares tend to outperform. You might think, “Hey look, this is a market that is so unique and country specific that only the investors in the country really outperform.” But we actually don’t see that. We see sophisticated investors globally coming in and earning alpha as well. All those things coupled together make it a really nice market to invest in when it comes from an alpha perspective.
Tobias: One of the research papers that you have written recently, I don’t think it’s published yet, is the– you’re just looking at factors that would be reasonably familiar to most US investors, and as they apply in China. Can you speak to them a little bit about what you see in China contrasting it with the US?
Vivek: Yes, absolutely. We actually have published a paper on factors a few years ago though we do have one specifically on the AH premium, which is a country specific factor within is that– [crosstalk]
Tobias: Is it after hours? What’s AH stand for?
A-Shares vs H-Shares
Vivek: Oh, A-shares and H-shares. That is A-shares listed onshore, H-shares listed offshore in Hong Kong. There are some firms that list in both. The list in Shanghai or Shenzhen, and they’ll also list in Hong Kong. You might think, well, it’s the same share. Indeed, these are dual listed. They’re not same voting rights, same dividend rights. You might think they probably are going to have roughly the same price. They do not have the same price at all.
The A-shares, on average, tend to be more expensive. This is primarily due to shorting restrictions as well as, let’s say, retail mania. Their preference for certain stocks is different than their preference of other. Like a state bank like Ping An has a very low AH premium, might even be negative. Whereas a firm like La Chapelle, which is a clothing company in China. The A-shares is actually much, much more expensive. It’s five times more expensive than the H-share.
Because Hong Kong investors tend to be more sophisticated, there is differential information there. You want to buy firms with low AH premiums and underweight firms with high AH premiums. If you’ve done this historically, you would tend to earn excess returns. There’s another way to attack it as well, which is you can just buy the H-share if the A-share is too expensive. There are advantages and disadvantages to that. That part, it depends on what you benchmark to and a bunch of other things. There are a variety of reasons for doing it. There are variety of reasons for not doing it.
Factors That Do Well, Continue To Do Well In China
Vivek: But speaking about factors in general, the most interesting thing– Actually, there are two interesting things about factors in China A-shares. First is, there have been about hundred plus factors discovered in the US. If you blindly apply those to China, you’ll earn a higher Sharpe ratio than you would had you implemented them in the US. It’s interesting for a couple reasons. One is that it tells you that the factors really work. These signals are actually predictive of return, because you’ve transplant into a market that’s totally different, and it works well. The fact that it works better tells you something else. It tells you that the market is more inefficient. There’s actually another aspect of China A-shares that’s also compelled. That is that if you look at the cross section of factor return—
So let’s say, you saw a factor do really well in China A-shares, and you saw another factor do really poorly in China A-shares, to what extent is that predictive of how it’s going to perform in the future? Is a factor that does really well is that likely to do well in the future? Is a factor that likely to do poorly in the future? Empirically finding in China, that is true.
You can rely on the correlation between let’s say, and this is kind of a rough way to estimate it. You divide the sample into 2000, 2010, 2011 to 2020. You run the correlation of returns of factors in the first period compared to factors in the second period. What you’re looking for there is, do high factor returns in the first period tend to correspond with high factor returns in the second period? The correlation you get there is about 0.6. In the US is about 0.2. It’s not there’s no information. This is about 0.18, I think, if I remember the number exactly. It’s not that there’s no information in factor returns in the US. It’s just that, it’s a lot more consistent in China, and that actually opens a couple of doors.
One is that you can calibrate specifically to China A-share returns. Yes, you should still use global priors absolutely. But when you fit your models, you can overweight China a little bit, because you know it’s going to be a little bit more consistent. The second thing you can do is you can find signals that are unique to China.
It’s not like you can’t do that in the US, it’s just that you lack the confidence. With factors that work globally, you can learn from the global performance and use that as a prior to learn about local performance. For example, momentum works everywhere, but one country. It’s likely the case that momentum still works in that country. It’s just the bad luck for that momentum in that country. But in China, there’s so much consistency there, that you can search out unique signals within China, and they’ll still be reasonably predictable going forward.
Foreign Holdings Are Released Daily In China
Vivek: I’ll just give a quick example there, because it’s such a short time period as well, it’s such a fascinating signal. I’ve mentioned that foreign investors outperform. One of the reasons why we know this is, in 2016, they’ve opened Hong Kong Connect, which allows global investors to access about 1300 China A-shares, the onshore China stocks, the ones listed in Shanghai and Shenzhen.
They also released the holdings– Every day, you can get this by the way. Every day, you can get the holdings that have foreign investors in these stocks, Connect investors in these stocks. What we find is that stocks with high Connect holdings tend to outperform those with low Connect holdings.
If a foreign investor likes his stock, that’s a good thing, if a foreign investor doesn’t like a stock, That’s a bad. That’s a short sample, 2016 to now. But it’s consistent. It is a consistent outperform. I think the information ratio is above 2.5 which is ridiculous, or whatever Sharpe ratio the factor. That’s a factor that performs the first year, second year, third year, you can consistently see the performance, and that’s something that you rarely find in other countries. You just don’t see stuff like that. The fact that we can find signals like that is incredible, and it’s suggestive of the inefficiencies of the market.
Tobias: There’s two questions that springs to mind at– The first one is, I think, I saw in your paper that– how does momentum– But let’s just start there. How does momentum perform in China?
Momentum Does Not Perform Well In China
Vivek: Momentum does not perform particularly well in in China, at least historically. Let me caveat this. We posed this paper in 2017 I believe. We were looking up until about 2016, and momentum didn’t perform well between 2000 and 2016. We scratched our chins, and we’re like, “There could be a couple things going on here.” It could be noise. Sometimes, something just happens to randomly underperform. It could be that China is different, and momentum just doesn’t work in China.
The investors are different. They don’t suffer from the anchoring effects of prospect theory. Subsequently, momentum did pretty well. This is why we want to anchor on global data a little bit. You want that prior to help inform the returns a little bit, but momentum in China is not like momentum in the US. Momentum in the US has done pretty well historically. Momentum in China is just sort of so-so. But you can elucidate that by seeing momentum’s performance globally comparing it to the terrible performance, I suppose, between 2000 and 2016 and say, look, it probably works okay, but it’s not going to be very good.
That’s a nuance that we want to hear. “Hey, a signal does or doesn’t work in a given country.” In China, we can be a little bit more confident that that is true. We don’t want to completely jettison everything we know about the rest of the world. You want to have that Bayesian prior there.
You can actually manifest it literally as a Bayesian prior. But there are other ways to do it. If you just don’t want to deal with the basic mathematics, you can just use regression predicting factor returns in China using both historical China factor returns and the global factor returns. What we find is that there’s a loading on both. It does prefer China. It overweights China relative to the rest of the world. It learns more from China, but it doesn’t mean the rest of the world data is completely useless.
Tobias: It’s interesting because AQR has that research that shows that Japan was one of the places that didn’t perform particularly well on a momentum basis, and I think that from recollection of people was something, it’s just likely that this is just luck, if you have enough sample periods. The other possibility is you’ve got a reasonably short period of time in China 2000 to 2016, and that period captures one of the worst drawdowns for the momentum factor in the US in like hundred something plus year, and it may be [crosstalk] only looked at that in the States, you’d find a similar kind of– What do you think about that?
Vivek: That’s an excellent point, and it was actually something that I really should mention. Momentum crashes. We know that in the US momentum had a massive crash at the inflection point of the financial crisis into the recovery. There was a similar thing that occurred in China, but they also have more peaks and crashes than other countries. In particular, in 2015, there was a massive run-up in stock prices, and a massive crash subsequently, and then the recovery after that. There was a momentum crasher that didn’t exist anywhere else, and it was because it had that inflection point like behavior. Momentum does work better. It is aided more by– the momentum has its moments, momentum crashes, implementations. It is aided more in China than it would be in other places. The reason why is because it does have more momentum crashes and you need to adjust your leverage up and down based on the beta that your long and your short portfolio have. You basically don’t want to be taking a lot of net beta risk that should make you want to lower your leverage there.
Tobias: It’s an interesting idea. I’m not sure what necessarily drives momentum, but I would have thought that if you had a market that was as heavily concentrated in retail as it tends to be and retail tends to be a little bit more chasing performance, I could see that being reasonably pronounced. If anything it’d be the other way around, but that’s not the case?
Vivek: Well, that’s such a fascinating question. The question goes to the root of momentum itself. I think the question is momentum, the overreaction itself, or is it the response to underreaction? Weirdly, I think empirically, I wish I could pull the papers out– recall the exact papers, but it’s both. It’s both an underreaction to historical information. Then, momentum is just the correct reaction. But then, it actually overcorrects, and then that component of momentum corrects over time, and that’s in part how we get long-term reversal. That’s such a fascinating aspect of momentum. That is a little bit of both. It’s a little bit of the underreaction and a little bit of the correction to the underreaction overreact.
The question then is, in China, is it the case that there is just more overreaction straight up? If this is more overreaction straight up, then the momentum effect will be dampened. I think it’s weak. Because it turns out that historically, if you look at the full period, momentum still works. But the reason it’s not as strong is probably because there’s less of the initial underreaction. They’re just the barreling towards the stock price that’s going up, and that’s amping up their inflows into the stock.
Value Performs Well In China
Tobias: Let’s get off momentum, because it’s not something I know a great deal about. Let’s go on to the fundamental factors that you look at. This is a value podcast for the most part, let’s talk a little bit about value factors, how have they performed in China?
Vivek: God, it’s such a– what’s the word, it is very volatile. Let’s put it that way. In 2016, 2017, 2018, value did very well. In 2019, in February 2019 in particular, there was a massive reversal. It was a rough turnaround in value. Back then, we were a little less diversified in terms of signals. We were a little bit more value actually than– Since then, it’s been a volatile ride, and interestingly, even this year, value did very well at the beginning of the year, and then it reversed, and then growth started outperforming year to date as of recently. I think yet again, it’s value starting to win. It’s a very volatile signal in China.
That being said, on average, it is very strong. Value tends to do very well in China, and it makes sense. This is a type of market, where value would tend to do well. It’s a retail-heavy market, it’s a market that is very story driven, very focused on the glamour stocks, so to speak. Signals that lean on value– the AH premium by the way doesn’t lean on value. There’s actually more going on to the AH premium than value, but a big part of it is just that, “Hey, the stock is cheaper in Hong Kong than it is in China.” So, value does do very well in in China, but it is a volatile signal, and you probably want other signals in there informing on that, so to speak. Yeah, go ahead.
Tobias: Well, there’s a couple of questions that come out of that. One is, does it largely follow the global experience with value? This year, for example, it started out fairly strongly for value, and then there was a reversal-
Tobias: -in Q2, and then it’s seems to have started up again really recently only over the last few weeks.
Tobias: Is that the experience in China as well?
Vivek: Interestingly year to date, it has followed that, but overall, it has not followed the value trajectory. Over the past decade, value has not done particularly well, depending on how you measure it. I think there have been some, let’s say, better ways to measure where value has done well between 2010 and 2018. But if you use a very traditional measure of value, you would not have done well for the past decade, you wouldn’t have done well over the past 20 years total.
If you just accumulated the past 20 years of data, value in the US hasn’t done well then, speaking from the HMO factor perspective, let’s say. But in China, it’s done well over the past decade. This is actually true about factors in general. In general, the China A-factors are uncorrelated. They have lower correlation with global factors. If we were to look at value in the US and develop XUS, I can’t pin down an exact value, but the correlation would be reasonably high. It wouldn’t be 0.2, which is what it would be in China A-shares versus the US. So, looking at value factors between China A and the US over long periods of time, very low correlation.
If you’re interested in getting another exposure to value that captures the same ideas, but doesn’t have a high correlation with global value, China A-shares is a great place. In fact, China A-shares, even the long-only market portfolio is not particularly correlated with global markets. I believe the correlation between, I’m sorry, US and emerging markets is something around 0.75 between US and China A-shares. It’s about 0.5. I would say about 0.55. That’s very low.
The correlation is even lower on the factor level. I’m sorry, I apologize. It’s lower than 0.5. It’s 0.45. My apologies. My memory is not serving me well. But it’s a nice feature. You don’t just want to diversify your beta, which is good. That’s important. It’s probably the most important thing. But diversifying your alpha source is also quite nice. The factor components, that those long short factors are very close to uncorrelated between the US and China A.
Tobias: You prefaced my question a little bit before when you referred to it, but the value factor is book to market, or price to book, or however you want to describe it. But clearly, there are other factors that have done much better in the States and globally. So, how are you thinking about value in a China context, and perhaps you could also talk about that in a sector or industry breakdown?
Stripping Out Value Components To Generate Better Returns
Vivek: For what it’s worth, and I think [unintelligible [00:27:45] Porter Stephens, there’s a very old paper in 2002, stuff like that. What they showed was that value and a variety of other signals work better industry neutral, so within the industry than betting across. Indeed, that is true in China as well. Frankly, it’s probably true across many signals. It’s fine to calculate the signal across all industries, across all stocks. But when you’re constraining your bets in your portfolio, you want to keep it relatively tight around industry.
But speaking to the different measures of value, indeed other measures have done well, and I have maybe a controversial opinion about why that is. I think the other signals do better, because they are leaning on other signals. Let me make sense of that. Price to book, book to market, that’s a prototypical value signal, hasn’t done particularly well over the past decade in the US.
But you look at earnings to price, does a little bit better. Cash flow to price, maybe a little bit better. Earnings to price, it is leaning on return on equity, leaning on a profitability metric. It gets a little bit from quality. Cash flow to price, that leans a little bit on accruals and profitability. If a firm is accruing a lot of earnings, it will have low cash flow versus earnings.
If they are earning a lot of their earnings through cash, then they’ll have a good price to cash flow or a low price to cash flow. If you use one of these other metrics like enterprise value instead of price, you’re shying away from firms with high debt. There are a variety of value metrics and some work better than others, but if you look at the component of return that’s driving that additional outperformance, it’s usually just coming from another signal. It’s coming from some other fundamental signal.
There are a couple ways you can handle that. One is, you can use the diversified set of value signals, and I think that’s perfectly sensible approach. The alternative is to use other signals that just directly capture that component that you’re getting. Use return on equity, use cash flow accruals, use leverage to capture high-debt firms on average and underperform. There are other ways to capture those, and if you’re using a sufficiently sophisticated model, it actually will barely matter.
It’ll pick up the information cleanly, extract the component that it needs, and predict using that. Especially if you’re doing a traditional factor approach, I would just capture the signal. I would just go after the additional single return on equity, go after leverage, go after interest expense, debt coverage ratio, whatever it happens to be. That’s where that additional return is coming from. Indeed, we do see that differential performance in China as well. But really empirically, it seems to be coming from those other signals, and just a correlation analysis will show it.
Tobias: Yeah, that’s really interesting. I’ve never heard that expressed before, but that makes perfect sense, that the flow metrics capture some quality, or accrual, or some other factor, and those factors have done quite well even through this period. If anything, they’ve done much better than they ordinarily do. One of the challenges, I guess, when you have somewhere like China, which is still reasonably new to this style of investment
Why Chinese Companies Don’t Like To Report Losses
Tobias: How do you tackle the challenge of, say, accounting?
Vivek: Yeah, oh, accounting is such a beautiful one. It really is. In China, there’s something different and unique that doesn’t quite exist anywhere else, and I’m glad you brought that up. In China– well, let me first speak about the US pre-2000, because it’s getting somewhere. Trust me, trust me. In the US, if you look to the histogram of return on equity pre-2000, what you would see is that it looks mostly like a normal distribution, and then at around zero, there was a small dip down on the negative end, and around on the positive side of zero, there’s a small push up.
What that is, is accounting manipulation. If you’re at negative one cent per share, then you’d say, “God, can we just adjust earnings a little bit, accrue a few more things here and there, and then get to positive one cent per share?” That’s used to happen. Actually, it doesn’t happen anymore by the way. If you look at the ROE graphing in the US post-2000, you don’t see that. SEC brought the hammer down, and that was the end of that.
In China, the regulators have not brought the hammer down at all. What you see is, you see a cliff function. It’s nearly flat at negative ROE, and then a huge spike. Then, looks like how it should look for the most part except for one thing. There’s no right tail. If you look at the highest ROEs in China A-shares, that’s missing too. Let’s think about how does all this fit together
Vivek: I’ll explain why they really don’t want to lose money a little bit later, but there’s an incentive not to lose money in China, to not show accounting losses. What they’re doing is, they’re trying to pick up as many negative accruals as they can, when earnings are high. That’s why you don’t see the right tail. They’re basically storing up negative accrual, so that they can take positive accruals if earnings are low.
But where they really capture the push into positive effect, because that gives you a little bit. But sometimes, you had a loss in a given year. So, how do you get into the positive territory? Extraordinary items. They’ll sell things. They’ll just have something, whatever division– Well, that division might be a little bit much, but they’ll have assets, and they’ll sell them off, and earn a particular profit on them, and that’ll be enough to push them into positive territory.
You see extraordinary items used a lot more by firms that are just in the negative earnings area and can just barely push themselves over. That’s why you see that cliff function, but then what’s the motivation? Why are they doing this? The reason is, if you lose money two years in a row in China, you are classified as a special treatment stock.
That just means your price limit goes from 10%. Your daily price on China is 10% for a normal stock. It goes up 10%, you can trade below that value, but no trades above 10%. It effectively shuts down the trading, same on the downward side. When you’re a special treatment stock, it’s 5%. Plus 5%, minus 5%. You might say that’s not a big deal, and it’s not. It’s not a huge deal. But there are a couple issues. One is many investment firms will just exclude those stocks altogether. Many investors, even retail investors, will just say, “God, I don’t want to hold that stock.”
Second, if you continue to lose money, you will get delisted, which is really bad. If you lose money for years in a row, that’s it. You’re off the exchange. That’s a difference in regulatory environment that then feeds into this fairly crazy accounting behavior. Behavior that doesn’t make a lot of sense outside of context of the threat of being delisted for losing money too many years in a row.
But what’s also interesting is the accounting rules themselves are actually very similar. It’s not like you sit there and you read the accounting rules, and you say, “Hey, the Chinese accounting standards are vastly different than international accounting standards.” That’s not the case. They’re very, very similar. They’re very much in line with the variation you would see across other countries. The manifestation is different. The manifestation comes from other regulatory rules, and just other firm-specific behaviors.
Tobias: Yeah, that’s really fascinating.
Trade Sizes Released Daily In China
Tobias: Anything else about the regulatory environment that’s unusual to China?
Vivek: I think the amount of releases. I’m trying to think of the most salient, punchy signal here. One of the interesting ones is the exchanges will release the trade sizes every day. Hey, there was this many buy orders in this size, this many sell orders in this size.
They’ll show the small net buys, the medium net buys, large net buys, extra-large enough buys. It turns out, as one might expect, when there are lot of large net buys on a given stock, that is a positive signal, because those are coming from two sources. They’re coming from institutions who are probably fairly informed, and they’re coming from– There’s actually some evidence of this that another researcher and again, forgive me, I do forget the author there, who found that those are probably coming from directors and people with some inside knowledge.
On the small side, those are retail investors. It turns out that the retail investors effectively have negative information. When they’re net buying something, that is a bad sign. When the large investors are net buying, that is a good sign.
Here’s an aspect of the regulatory environment where they are releasing information or they’re forcing firms to release information, exchanges, I should say. Then, investors who can then leverage that information to predict returns.
That’s an aspect of the regulatory environment that’s actually quite nice. There’s another aspect of the regulatory environment, which is rules change fairly quickly, and I don’t want to call it– They’re not arbitrary. I genuinely don’t believe they’re arbitrary. I’m not trying to be diplomatic here. There is sense to them. You just don’t know the sense until you’re exposed. Sometimes, you’re surprised. You’re like, “Whoa, but I didn’t expect that.”
How China Cut-Off Short Selling
Vivek: In 2015, they effectively cut off short selling. The way they did that is by telling brokers– Well, first of all, they punish what they call some malicious short sellers. This was during that big crash in 2015. Market went up, crashed. They then blamed malicious short sellers, and they highly encouraged brokers to not lend shares and not allow people to lend shares. That basically stopped short selling for quite a while.
They’ve only recently started reintroducing it. There’s actually quite a spike. Short selling is accelerating quite a bit. But you’ll see a lot of these changes. They’re still figuring things out. But it’s a fairly young market, the Shanghai and Shenzhen exchanges were founded in 1990, 1991 respectively. These are not old markets. They’re quite young. The regulator’s still feeling things out a little bit. You see that manifest in the rules. There are somewhat frequent rule changes.
But one benefit is that because they’re risk averse, they require a lot of disclosures. They want firms to disclose a lot of information, they want exchanges to disclose a lot of information. That’s how they feel that– They want to protect their investments. The China A-shares investors are generally within China. On average, the ownership is not foreign, it’s within the borders, and so they care a lot about protecting those investors and so force a lot of disclosures, those disclosures are extremely valuable. They provide so much information, and so that aspect of the regulatory environment is quite nice.
China’s SOE Reforms
Tobias: What about the impact of state ownership?
Vivek: Yes, yes, absolutely. State-owned firms, it’s the one of the most difficult things to describe, because it historically, state-owned firms have underperformed. We have it as a signal in our model. Recently, there’s been state owned– There’s been SOE reforms. They’re trying to make these firms more profitable, less zombie-like, and we have to make a decision. We are keeping the signal as of now. We’re using it as a predictive mechanism.
We need to make a decision to figure out whether the signal should still be in there or not, because if you’re looking at it historically, you would say, “Look, state-owned firms, their profit growth is worse. They have underperformed historically. You’d generally want to be a non-SOEs than SOEs.” That has historically been the case.
But if SOE reform is sufficiently successful, that differential will not be as strong. So, the model will make a mistake. It will think history will repeat itself, but it won’t. That’s a difficult question. I wish I had a firm answer like, “Yes, I am certain that it will go one way or the other,” but I’m not. It is still in the model. I would taper my expectations in the [unintelligible 00:44:42]. I don’t think the underperformance of SOEs historically will be of the same magnitude. I think it is unlikely.
Tobias: Just one final question about the fund. It’s Quantamental, and so that– Can you just talk a little bit about what does ‘quantamental’ mean, and then what is the impact on the fund, or what does that mean in a practical sense?
Vivek: Yes, absolutely. Quantamental means a lot of things to a lot of people. I think on one extreme, you have quant inputs to a fundamental choice. That’s one side of quantamental. Then, the other side of quantamental, is, you have some fundamental signals in your fund. We’re somewhere in the middle, and in particular, we have a fundamental team in China that learns about signals, discovers signals, speaks to people on the ground, understands the unique disclosures within China, and helps us build signals out of that, and then those go into our quant model.
Then, when we review the weights– we’ll just trade the weights that are generated, but when we review those weights, we’ll sometimes see something that just doesn’t quite jive with the intuitions of the fundamental thing. What we do there is we say, “What signal is missing from the model that will help elucidate that idea?”
There was an analyst report that came out that’s actually very negative on this company. Okay, great. We need to capture analysts report, signals, and indeed we do. There’s negative news on this company, and that’s our capture model. We should encode the news, and then put that into the– capture the news segment and put that into a model.
That’s our brand of quantamental. It’s that we have this local team providing local insights to the model, basically giving us new signals. We test those signals in a quant framework, and add them to the model, and the fundamental team will then look at the weights, evaluate the weights, and figure out what might we be missing given that we’re making these bets. So, that’s the full range of our quantamental processing.
Tobias: Yeah, it’s absolutely fascinating, Vivek. If folks want to get in contact with you or follow along with what you’re doing, how do they go about doing that?
Vivek: I think LinkedIn is the best way. If you could put my LinkedIn URL there, that’d be great.
Tobias: Yeah, well, Vivek Viswanathan, thank you very much for your time.
Vivek: Wonderful. Thank you.
For all the latest news and podcasts, join our free newsletter here.
Don’t forget to check out our FREE Large Cap 1000 – Stock Screener, here at The Acquirer’s Multiple: