Q2 2024: 24.58% YTD, the case against microcaps and MTR (HK's rapid transit system)

Disclaimer: Nothing on this site, including this page, constitutes investing advice. Please do your own due diligence.

Q2 YTD performance is 24.58%. This is slightly ahead of the Nasdaq 100 (~19%) and ahead of most other indices.

What is actually knowable?

There are hard limits on cases where knowledge and hard work is useful. As a silly example, humans have worked hard, discovered math and science, religion, philosophy, tripped on drugs, and so on. After thousands of years of wondering, do we know with complete conviction, what is the exact nature of the universe? In this question, there cannot be any degree of precision regardless of the centuries of hard work. There are many practical cases where this is true, both in investing and elsewhere. As an example, someone can know the oil & gas industry inside and out. They can spend a lifetime understanding every nut and bolt that goes into an oil rig. But because the world is highly uncertain (we have wars, pandemics, etc.) it is basically impossible to know what the price of WTI crude will be in 10 years. 

There are other pretty hard limits to what an investor can know. In a previous life, I have worked at a couple of Nasdaq 100 companies. Whenever I read a professional analyst's reports on any of these companies, I find wild inaccuracies despite it being their job to know. The reality is, it’s basically impossible for analysts and investors to deeply know what goes on day to day in a prospective company unless they themselves were former employees. An employee is completely immersed and might be familiar with the personalities/incentives of all major executives, hundreds of colleagues, progress of all major projects, how impactful they are, etc. To compound the problem, I see management teams regularly pull the wool over investors eyes selling buzzwords that have no meaning but that investors want to hear (think blockchain, AI, etc.). It’s just incredibly hard for someone on the outside looking in to really know what’s going on. And yet, despite the magnitude of difficulty, many professional analysts seem to talk confidently about the nature of these businesses. I’d argue the world of professional investing is full of people who confidently feel they know the companies they own, but actually have next to no clue.

A few thoughts given the above:

• It is better to focus on knowledge that is within the bounds of usefulness in the long term
• It is of paramount importance to be aware how little anyone can possibly know, there’s a premium to be placed on information that is verified to be true in a world of false information
• It is better to focus on investment cases where the core facts which imply an outcome are actually knowable
• It is important to distinguish between work that leads to knowable conclusions or if we’re just doing work that gives us false confidence

As a practical example, a lot of investing in inflection tends to forgive past mistakes in favour of a glorious future. Management's previous mistakes are knowable. Oftentimes, the future is not. As a result, personally, I tend to be incredibly unforgiving to management teams that have displayed a certain degree of stupidity even when selling a future that implies significant upside.

The priority queue and the case against microcaps

One of Ben Graham’s (Buffett’s teacher and father to value investing) famous quotes is “in the short run, the market is a voting machine, but in the long run, it is a weighing machine.” This is commonly interpreted by investors to focus on the long term and ignore the short term. However, I do not think it is completely futile to attempt to have a mental model for how the voting machine works. Say, as an example, there are two equities, both equally undervalued. One equity is a microcap (ie. small company) and the other is a large cap (ie. large company). Which company is likely to be bid up by investors first? Since there are fewer large companies than small companies and there are only so many vehicles that large funds can invest in, large caps are more visible, it’s not hard to imagine the larger company is bid up first. Microcaps, on the underhand, can remain undervalued for years.

In computer science, there’s a very common data structure called the priority queue. This data structure allows algorithms to enqueue items in any order and dequeue those items based on priority. That priority is typically determined by an objective function which scores items. As an example, say items are added (enqueued) in order to a priority queue with scores 5, 10, 2, and 1. The items would be taken out of (dequeued) the queue in the order of 10, 5, 2, and 1. If we visualise undervalued securities as a priority queue where high priority securities realise their intrinsic value before low priority securities, what would likely be the objective function? 

I imagine the objective function would weigh priority based on some the following criteria:

1. Visibility
2. Certainty
3. Magnitude of new information/events
4. Magnitude of mispricing
5. Relevance to the current zeitgeist (ie. at the time of writing, AI stocks are hot)
6. Etc.

At the top of the list is visibility. After all, if nobody knows about a mispriced security, how is it possible for it to be bid up? The logic for investing in microcaps goes something like: mispricing in micro caps is probably going to be larger than with larger equities because they are less visible, they have less analyst coverage, etc. As a result, I can discover these equities before the big fish can invest and my upside is larger. 

However, microcap investors have been waiting for an awful long time and I’d argue they might be waiting for a long time to come. However, unless one wants to wait forever, part of the investment criteria after fundamental value should probably be the criteria above (visibility, certainty, etc.). I’d argue, if there is obvious value to be found in plain sight (ie. in large caps), this should take priority over value that is hidden deep in the haystack (ie. in microcaps). That is, the objective function that I will use to score investment opportunities will mirror what I believe to be the market’s objective function. 

MTR Corporation (66.hk)

As an example, MTR Corporation’s investment thesis has elements of both of the above (knowability and priority queue criteria). MTR is many things but it is principally known as the rapid transit train in Hong Kong. MTR carries 4.6M daily passenger rides daily for a population of 7.346M. To say MTR is vital to the functioning of Hong Kong would be an understatement. Unlike many municipal rapid transit systems, MTR is profitable on fares alone partially because of the density of Hong Kong and ubiquity of use. The density is a feature both for the utility of a train rapid transit system as well as detractor for the competition, cars. Cars require parking and congestion can plague central areas of Hong Kong. 

MTR is trading at levels not seen in over a decade, both on a price and valuation perspective. Why? A few things. Firstly, MTR is planning to spend 87.9B HKD between 2024-2026. 2023 EBITDA was 14.9B. Essentially MTR is planning to spend 5.9x 2023 EBITDA in a span of 3 years. Secondly, EBITDA from property development took a hit going from 11.6B in 2022 to 2.3B in 2023B. MTR had a tender for a development site at Tung Chung which received no acceptable bids. So cash flow coming from property development might look bleak. Anecdotally, many retail spaces are completely empty in MTR stations where that space could have been collecting rental revenue. Finally, as the HK government is a majority owner of MTR, it can be argued that there’s a conflict of interest between HK SAR and minority shareholders. 

Optically, I feel 2023 numbers distort the normalised earning power of MTR. 2024 is the first year in 3 years where net immigration into HK is positive. MTR should also be booking revenue from The Southside, Ho Man Tin, and Lohas Park developments. Although ROE for MTR is typically quite low, MTR is trading at a historic 0.86x book where the mean valuation over the last 10 years is 1.46x. Finally, we’ve witnessed spectacular performance of government controlled entities in recent history (ie. CNOOC being a 2x). They have been increasingly demonstrating shareholder friendly policies as China wants to be an investment destination.

The one difficult area of contention, in my mind, is whether new capex is going to be effectively spent. As I am of the belief that:

1. Hong Kong still has a long line of people, especially from the Chinese mainland, who would like to immigrate to HK
2. Despite the recent downturn in Hong Kong real estate, HK real estate is still very expensive (price to income ratio of 18.1 where 3 is considered affordable)
3. Real estate next to MTR stations is highly desirable

I tend to believe that while there may be short term challenges, there will be demand for both MTR stations as well as new developments hitting the market in the long term. 

Net-net: Hong Kong is returning back to normal, 2024 is likely going to look like a much better year from an earnings standpoint, and we can buy a monopoly business for a historic low valuation. By comparison, CLSA has MTR trading at an 6.6x NTM earnings which seems exceptionally low for a monopoly business.