An atypical investment approach based on insurance principles

‘We’ve built a methodology completely around the idea of avoiding losers’: Julian Koski, co-founder and CIO, New Age Alpha.

RYK VAN NIEKERK: Welcome to this market commentator podcast. My name is Ryk van Niekerk and my guest today is Julian Koski. He is the co-founder and chief investment officer of New Age Alpha. New Age Alpha was founded in 2018, and its headquarters are in Rye in New York. The company also has an office in South Africa in Greenside, here in Johannesburg. And even more interesting is that Julian Koski built and owns the Kubili House Safari lodge near the Kruger National Park. Julian was born in South Africa, but left for the US in the early 1980s. Julian, thank you so much for joining me. It seems that you can leave South Africa, but South Africa never leaves you.

JULIAN KOSKI: Absolutely, absolutely. South Africa has never lost me, and I love coming back there, and I come back there very, very often because I really love going there. South Africa and New York are my two loves and it’s where I spend my time.

RYK VAN NIEKERK: There are definitely not many asset managers who have an office primarily in the US and also in South Africa. How big is the South African part of the operation?

JULIAN KOSKI: The South African part of the operation is really a sub-advised relationship that we have with global and local there in Johannesburg. So it’s a big part of what we do, but the majority of what we do is operationally done out of the United States.

RYK VAN NIEKERK: The name of the business is New Age Alpha, and I always find it very interesting when an asset manager uses Alpha in its name. It can come back and bite you a bit. But tell us about what exactly New Age Alpha is, and what your investment methodology is.

JULIAN KOSKI: Sure. Well, it all starts with a basic philosophy that we have that we think portfolio managers have an imperfect understanding of where ‘alpha actually comes from in the first place. They think in terms of picking winning stocks, when really the goal should be to avoid losers. Right?

If I drew a straight line for you, and that straight line was the S&P 500, and I put winners at one end and the losers at the other, and I asked an audience of portfolio managers how they would design a portfolio to beat the S&P 500, 99% pick the winners. But to pick a winner, you have to have some knowledge of the future. The future by definition is not known. And the more we actually forecast this unknown future, the more we are increasing the likelihood of being wrong, and investing in a loser.

What we’ve done is we’ve built a methodology completely around the idea of avoiding losers, and we’ve departed from traditional portfolio management ideas. In fact, we’ve drawn on the principles of insurance to do this.

RYK VAN NIEKERK: That’s a very interesting approach. But avoiding losers – is that not maybe a typical focus area for hedge funds, to maybe rather short the losers than go long on the winners, or what you perceive to be winners?

JULIAN KOSKI: Well, in fact, what we notice about this risk [is that] there is a risk inside stock prices that is different from firm-specific risks. So with the risk that a factory burns down or a CEO passes away, the traditional accepted notion is you diversify your portfolio as a way to mitigate that risk.

But this is a risk that affects stock prices that investors aren’t aware of. They’re not even getting paid for taking this risk. But, most importantly, you can’t diversify this risk away. Normally the higher the risk, the higher the return. With this particular risk, the higher the risk it’s not getting a better return. So the only way of dealing with this risk is to actually avoid this risk. It’s not about trying to mitigate the risk by trading the risk, because stocks that have a lot of human bias built into them, those are not stocks you actually want to essentially short. You can, but it’s better to avoid them.

If you watch what happened with GameStop [on Thursday], here in the United States, that stock has got, we calculate mathematically, a 100% chance. It will fail to deliver the growth implied by its stock price. So yes, you can short that stock. But look what could happen if the rest of the behaviour catches up with you, you might lose a lot of money. Just ask the hedge funds what’s happening to them.

RYK VAN NIEKERK: You’ve recently launched the New Age Alpha US large-cap Equity Avoider ETF. Can you maybe just explain exactly how it works and which shares are included in that ETF?

JULIAN KOSKI: Basically we’ve launched two.

Read the fund fact sheets:
AVDR US LargeCap Leading ETF

The first one is called AVDR. What that is, we take the S&P 500 and essentially we remove 450 names, the companies that have got the greatest chance of failing to deliver the growth implied by that stock price. And, as I said, we use a mathematical formula to come up with that; we are left with 50 names that remain in the portfolio, and we weight them by our probability. In other words, we provide more ownership of those names that have got the lowest age-factor scores by way of probability.

So in terms of the names and the things we actually own, I can give you some names if you’re interested. If you just look at the holdings of these companies. They are generally names you associate with S&P 500 ownership, but they are Occidental Petroleum, Autodesk, Broadcom, MSCI, the two probably well-known names, Goldman Sachs and Walt Disney, Darden Restaurants, and Teradyne. Those are the top 10 names.

RYK VAN NIEKERK: And the Fangs – the Facebooks, Alphabets and the like – are they included?

JULIAN KOSKI: No, they’re not. They’re not for a very simple reason. Remember, we don’t talk about failure because we think Amazon and Facebook and those will [fail]. No, what we’re saying is there’s another way to generate alpha without owning those names, because if any one of those names fail to deliver the growth implied by their stock price, they’re not going to go out of business, but they’re going to hurt someone’s portfolio. So all the alpha or outperformance you’ve created will get eroded by the fact that you own those names. And those names today, as a percentage of the S&P 500, really will hurt the portfolio.

So is there another way of having S&P 500 returns that can beat the S&P 500 without the ownership of the Fangs? The answer is yes.

RYK VAN NIEKERK: The fund was launched quite recently, but did you prospectively look at what the returns would have been if you’d started earlier and used the same methodology?

JULIAN KOSKI: Oh yes, absolutely. The indexes have been live for a very long time, so we’ve got the track record of the indexes. So, for instance, last year the S&P 500 was up 18.4%. This was up 22.84%. And that’s what we expect to happen. That’s why we call it New Age Alpha. We believe there’s a new way of generating alpha that is radically different from what people do today. We think the investors that actually think that they’re generating alpha are contributing to the problem – and we are disintermediating that problem.

RYK VAN NIEKERK: I’m looking at your website and I see you actually list a significant number of indices with performances. Are there exchange-traded funds [ETFs] linked to all of these indices?

JULIAN KOSKI: There are only two right now. We’ve done the S&P 500 one, and we’ve done an ESG fund, but this year we’ll be launching another 10 to 12 different ETFs. Yes. The goal is that we’ll be launching a family of products around this methodology. We’ve just rolled out the first two and we’ll be doing probably another 10 sometime this year.

RYK VAN NIEKERK: What are the risk profiles relative to the risk profiles of markets at the moment, because we are seeing equity markets running significantly. And while we are seeing depressed economic prospects, we’ve seen a lot of liquidity in the market and a lot of expectation; but there’s a suggestion that the risk profile is higher than it was in the past. Do these portfolios protect you against that risk?

JULIAN KOSKI: Yes. I’ll give you an example of that 50-stock component portfolio of the S&P 500. It has the same standard deviation of 14.9 as the S&P 500. So you’re getting a better return with the same risk profile, but you’re getting 112% of the upside capture, and 86% of the down. So the goal, yes, is to protect the investor in the up and the down. It’s not a hedge fund. It’s not going to be able to protect you on the short side, but it’s got a much better risk profile in terms of upside capture and downside capture, with the same standard deviation and getting a better return. There just can’t be 500 stocks that can all contribute to the returns. There are losers in there, and you want to remove them.

RYK VAN NIEKERK: Have you seen an inflow of funds since launch?

JULIAN KOSKI: Yes, just beginning now. We’ve got institutional investors, so we are owned half by an institution. Their capital is coming in and we’re beginning – the sales process is beginning. So flows are coming in daily now. They’re all coming in daily.

RYK VAN NIEKERK: And are they available to buy in South Africa?

JULIAN KOSKI: I think they are. I think it’s just a brokerage account. Are South Africans allowed to trade USA equities? I’m not sure. The products in South Africa are designed for South Africans. Those are not ETFs. Those are unit trusts that are being offered

RYK VAN NIEKERK: Yes, you can if you open the correct trading account. What do you think of market valuations at the moment, and how do you think investors should view these valuations?

JULIAN KOSKI: Again, we don’t forecast the future. But what our indicators are saying is that the S&P is reaching a point of fair value in our view. Right? So if you look at the math and you do the math behind the S&P 500, you’ve got a majority of firms coalescing around the 50% probability of delivering growth, or failing to deliver growth implied by experts, which would suggest fair valuation if it’s at 50/50. But then you do have names that have much lower probabilities of failing.

So it is the time to be a stock picker, because you want to be careful of the ones that are fairly valued, and you really want to focus on the ones that are overvalued and undervalued. So yes, in general the market’s fairly valued, but there’s still a lot of opportunity. You’re probably not going to find it within the S&P 500 as a whole, but you will find it within a portfolio of stocks that essentially either picks or avoids stocks around the S&P 500.

RYK VAN NIEKERK: I think it’s a very interesting approach. You have your contrarian investors or your deep-value investors who would actually go and look at those losers and see whether there could be recoveries. That actually just shows you that there are many, many different methodologies out there, and I don’t think many investors actually go and look at the methodologies and understand exactly what the differences are.

JULIAN KOSKI: Well, there’s one notion I remember, growing up in South Africa – it’s that insurance companies own the buildings that banks are in. And there’s a reason, because they look at risk completely differently. We have a saying – ‘Manage risk like an actuary, not like a portfolio manager’. What we mean by that is, if you think about an actuary, does he ask you: ‘Are you going to quit smoking, are you’re going to go to the gym?’ No, he could never underwrite risk based on that kind of vagueness. He uses mathematics to do it. The same thing here.

When you think about a stock, there are only two things absolutely known about a stock at any point in time – and that is the actual stock price and the financial statements. The rest is all vague and ambiguous information. And that is exacerbated by portfolio managers, because they impound that information into those prices, and they make the situation worse, far worse.

RYK VAN NIEKERK: Who developed your mathematical model?

JULIAN KOSKI: My partner and I together developed it. So we, myself and Armen Arus, the other co-founder. Armen Arus is the CEO of the company. Together we’ve done it. We’ve worked together over the previous 20 years doing this. This is a long process of development.

We built another company, Transparent Value, that we sold to Guggenheim Partners in 2015. This is the next phase of our lives, and we’re doing it again.

RYK VAN NIEKERK: I think that, especially in the crypto space, there are many institutions or individuals who claim they’ve developed bots that are trading at 1% margins every single day. Can you actually predict valuations and market movements through a mathematical system or methodology? How much human intervention is still needed to select the correct equities or the incorrect holdings?

JULIAN KOSKI: On our side it’s zero involvement by humans. It’s all math. So we’re using mathematics. Let me give you maybe a quick example of the math, the way it works.

Let’s take Tesla for a moment. Just look at Tesla. If you took a stock like Tesla – let’s say it was trading at $100 pershare – that share price implies that they have to sell a certain number of cars every quarter. Now let’s say that that implied that they had to sell a thousand cars per quarter. Now, if you open their financial statements for the prior 16 quarters, and you calculated a historical distribution of that growth rate, one could safely assume that, in the next quarter – had they done a thousand cars per quarter for the prior 16 quarters – in the next quarter there’s about a 100% chance they’re going to do it again.

But now, if that stock price is at $500 a share, and let’s say that implies that they now sell 10 000 cars per quarter, well, guess what? If in the prior 16 quarters they’d only done a thousand, the probability in the next quarter of selling 10 000 cars is about 5%. So the more the stock price is impacted by human behaviour, the more the company has to deliver to support that growth.

And a lot of companies historically might not have done that. There’s a high chance of failure. So it’s actually a very simple algorithm. It just looks to see, wait a minute, that’s what that price is doing. What does it imply? And therefore, based on the past, can it actually deliver that? And, trust me, there’s a lot of hover around the simplicity of that. It’s just understanding what the price has done.

RYK VAN NIEKERK: Yes, I understand what you’re saying, but virtually every single investor in Tesla is smiling at the moment, because the performance has been absolutely phenomenal.

JULIAN KOSKI: And they should have been smiling because, if you did this math – I did this math 18 months ago – Tesla was trading at $253 a share 18 months ago. Now that implied that Tesla generates $6 billion in revenue. Furthermore, it implied that they sell 95 000 new cars per quarter to justify their price.

Now, I did the exercise. Look back 16 quarters, and what you saw was that Tesla had an 80% chance to deliver that growth. Well, guess what? That meant the company was completely undervalued.

Let’s just go back in time on that date, when it was trading at $253 per share: there were 169 headlines about Tesla out there; 130 of those headlines were negative. You had some of the biggest shorts in the market saying Tesla was going to go out of business. But that’s the very vague and ambiguous information I’m talking about. It found its way into the stock price, and in that case absolutely underpriced Tesla. And one can agree – Tesla was completely underpriced. So that’s what we did. What you’re saying is right, Ryk. People made a lot of money and deserved to make a lot of money. It was the right goal.

RYK VAN NIEKERK: Just seems counter-intuitive that a company that has run so hard is included in a risk-averse portfolio like yours.

JULIAN KOSKI: No. You see, when we had Tesla back then, we owned it then. But we don’t own it now because now Tesla’s probability of delivering the growth implied by the price is 50%. It’s reaching fair value. There’s no opportunity there. But back then, absolutely. It was a screaming buy, because there was literally an 80% chance that they were going to deliver. So, based on odds, if you think about what we’re doing, we’re calculating the mispriced odds that are in the market. It’s almost like horse racing. You’re looking to see where the odds are mispriced from human behaviour, and you are buying the stocks that are mispriced.

RYK VAN NIEKERK: It’s very interesting. How often do you change the composition of the portfolio?

JULIAN KOSKI: Every quarter we do it because every quarter there are new financial statements that come up. So the models are updated every day for the current stock price, but then quarterly for the financial statements that come out. And that’s all that goes into it.

RYK VAN NIEKERK: How many mathematical inputs do you use to look at a specific company?

JULIAN KOSKI: I don’t know the number offhand. It’s not considerable, but think of a balance sheet and income statement, what’s in there. Maybe it’s worth looking at how things are done today. The way things are done today is an analyst will sit down and he’ll make some kind of forecast or assumption about the future. He is going to predict how many cars Tesla’s going to sell. And from that outlook, the outcome of that’s going to be if he thinks the stock is under- or overvalued.

We start with the stock price. So think of the stock prices an input, the primary input. And then you take your balance sheet and your income statement and you work backwards to calculate what that growth rate implies. Once you’ve got that number, you then look to see, well, what’s the likelihood they’ll deliver that number based on prior performance. So the inputs – I don’t know the exact number, Ryk , but it’s whatever the balance sheet and income statement would hold.

RYK VAN NIEKERK: Interesting. Julian, thank you so much for your insights and time today. This is an almost unique type of investment approach, and good luck for the future.

JULIAN KOSKI: Thank you very much, Ryk, much appreciated.

RYK VAN NIEKERK: That was Julian Koski, the co-founder and chief investment officer of New Age Alpha.



Sort by:
  • Oldest first
  • Newest first
  • Top voted

You must be signed in and an Insider Gold subscriber to comment.



Unfortunately multi-managers and manager researchers will take a long time to understand this.

It smells too much of actuary, no thanks

End of comments.



Subscribe to our mailing list

* indicates required
Moneyweb newsletters

Instrument Details  

You do not have any portfolios, please create one here.
You do not have an alert portfolio, please create one here.

Follow us:

Search Articles:
Click a Company: