Mark to market: Should investors jump on this new ETF offering?
Mark answers a reader question about a growth at reasonable price investment approach and the new GARP ETF.
Mentioned: Global X Management (AUS) Ltd (GARP)
Question:
Good afternoon Mark
Could you please do a deep dive/your thoughts on this factor-based strategy on the podcast?
Official document: Indexing GARP Strategies: A Practitioner's Guide
Peter Lynch apparently used this strategy to achieve 29% annual return over 13 years. Absolutely crazy!
Currently in Australia, I think Global X is the first one to introduce this ETF (released in September). Other big houses like iShares, Vanguard and SPDR may have their Garp-based ETF’s in the US but not here (yet).
Answer:
Great question. We can start with some definitions. GARP stands for growth at a reasonable price. The notion is that you want to buy shares that grow but you don’t want to pay too much for them. I don’t want to oversimplify something in an industry that thrives on complexity but if I had to rename this strategy I would call it…..investing.
GARP is what every investor wants. You want to own companies that grow. But you don’t want to pay too much for them. What are the alternatives? GARV – growth at ridiculous valuations? MPSC – moderately priced shrinking companies?
The question mentioned Peter Lynch. And Peter Lynch is one of the greatest investors of all time. He managed Fidelity's Magellan Fund from 1977 to 1990. His annual average return was 29.2% a year. That is shocking.
An interesting fact about Peter Lynch is that he got his first job at Fidelity after caddying for Fidelity’s president. He must have recommended some good clubs. This wasn’t exactly a Horatio Alger tale since he had just gotten his MBA from the University of Pennsylvania Wharton School of Business which makes him qualified on paper. But a nice story nonetheless.
Peter Lynch did popularise the notion of a GARP approach. He wrote a famous investing book called One up on Wall Street describing his technique. It is important to be clear that he was an active manager. Those returns were from him. And while I understand that there is a benefit from associating the strategy with Peter Lynch he will not be managing your ETF.
Professional investors like Peter Lynch and our equity analysts at Morningstar generally use some variation of a discounted cash flow model to value a share. A discounted cash flow is simple to explain and hard to do. The value of any company is based on future performance. Good performance for a company is the size of the cash flows that are generated. Since the cash flows occur in the future and we are trying to come up with a value today we need to discount them.
That process gives you a valuation or range of valuations that can be compared to the current share price. The model captures estimated growth in the projection of future cash flows and compares the estimated value to the current share price. Shares that are undervalued and expected to grow could be described as growth at a reasonable price. That is why I think GARP is just another name for investing.
The devil is in the details. How do you accurately estimate something that occurs in an unknowable future? How do you pick an appropriate discount rate? That is the part that is really hard. Peter Lynch did that very well. Many people don’t.
The question is how to replicate a strategy that is open to so much interpretation in a factor ETF which uses data points to pick shares. What is a reasonable amount to pay for a certain level of growth?
Global X S&P World ex Australia GARP ETF
As the question pointed out Global X just released a new factor ETF that tries to replicate this strategy. Appropriately the ticker symbol is GARP. Our analysts do not rate this ETF. These are just my thoughts on what an investor should think about when considering it.
This is a factor ETF. Factor investing involves the identification of certain characteristics in shares that have historically led to positive outcomes for investors. Once the desired characteristic is picked shares are screened to find individual companies that have those attributes.
The ETF tracks the S&P WORLD EX-AUSTRALIA GARP INDEX. To state the obvious these are global shares. The index ‘measures the performance of the top 250 growth stocks with high quality and value composite scores within the S&P Developed LargeMidCap Index (the underlying index), excluding stocks from Korea and Australia.’ I don’t know why Korea is excluded although I’m sure there is a good reason. I’m not losing any sleep over it. I don’t think you should either.
The ETF combines shares that score high on two growth factors – three-year EPS growth and three-year sales per share growth – and three quality and value factors – financial leverage ratio, return on equity and earnings to price ratio.
How are they combined? Well sometimes the index methodology document explains it so clearly no further explanation is needed. In this case the explanation is as follows:
“Calculate the Growth z-score as the winsorized z-score average of two factors: three-year EPS growth and three-year SPS growth. If the z-score for one factor cannot be properly calculated, use the z-score of the other factor as the growth z-score.
Calculate the QV composite z-score as the winsorized z-score average of three factors: financial leverage ratio, return on equity, and earnings to price ratio. A stock must have at least one of the quality factors and the value factor to calculate the QV composite score. If the z-score for one of the quality scores can’t be properly calculated, use the z-score of the other quality factor.”
Got it? Me neither. A z-score is a statistical measurement of a score's relationship to the mean in a group of scores. Winsorized is the transformation of statistics by limiting extreme values in the statistical data to reduce the effect of possibly spurious outliers. I would describe this as ranking shares against the average score and eliminating outliers.
Once the ranking is complete shares are selected that score highly in both categories and some restrictions on exposure to individual shares and sectors is applied.
Reading the selection criteria I’m picturing Peter Lynch sitting in his office in 1977. Perhaps he is thinking about his famous maxim “buy what you know.” He is suddenly inspired by looking at the Coke he recently purchased. Off goes an assistant to a building sized mainframe with less computing power than my first iPhone. The assistant has one mission – calculate the z-score on the chain store that sold Peter Lynch his Coke.
I’m obviously being a bit facetious here. There are things that I really like about this approach. I am a fan of all the criteria that is being used to select shares. This is ultimately just doing what all investors do. It is taking the only thing we know which is historic results and trying to use them to estimate what will happen in the future.
This is not a human picking shares where success is impacted by intelligence, knowledge, experience, biases and countless other things. It is taking a combination of factors and buying a widely diversified portfolio of shares that meet them. This diversification provides protection from the factor model picking an individual share that looks good from a historical perspective but has poor future prospects.
There is something romantic about a genius sitting in an office and finding the best shares to beat the market. But the data shows this doesn’t work very well. In the latest Morningstar Active / Passive Barometer Report over the last 10 years only 19.70% of active managers beat the benchmark in the US Large-Cap Blend category which is a reasonable approximation for an ETF with slightly more than 70% of assets in US shares.
The result of the process that GARP uses to pick shares is interesting. The following chart shows the top 10 holdings.
In some senses this looks a bit like the S&P 500. All the usual suspects are there with Tesla, Apple, NVIDIA, META, Microsoft and Alphabet. There are some interesting picks with oil giants Exxon and Chevron.
The ETF is new and there isn’t much performance history. However, the index has been around for a while and over the last 5 years it has outperformed the MSCI World Ex Australia index which is tracked by Vanguard MSCI Intl ETF VGS by 5.50% per year. I do want to point out that an ETF has a fee and the performance of the ETF reflects those fees. An index has no fee. In this case the fee is reasonable for a factor ETF at 0.30%.
Since the ETF is new I couldn’t find data on portfolio turnover which represents how often the ETF changes the holdings. This can have an impact on after-tax returns since capital gains can be generated when some shares are sold and others are purchased. It is something to watch. Twice a year the factor model will be re-run and if new shares rank higher on the criteria than existing shares the ETF will switch up holdings.
As always, the key for any investor considering this ETF is if the approach aligns with your goals and investment strategy.
If you have a question you would like answered please email me at [email protected]
