Why does sector coverage impact success, despite market-neutral strategy?

Investment Analyst in HF - EquityHedge

Let me reframe the question ... why would a market-neutral strategy do better just because the sector outperformed? It's 50% short still.

Because the sector outperformed in other ways such as momentum (winners keep beat & raise) and already has high beta. Being delta $ neutral doesn't mean you are necessarily beta or momo neutral. So even the market neutral funds did well. If APP rips 30% two months in a row while your MAG7 funding short is only up 10%, even if Citadel risk central forces you to degross APP back down every week, you're still generating a ton of uncompounded alpha. 

The real answer is that TMT stocks have cross sector momentum. That means there is not only large spread to capture but it can be done in short periods of time. This means you get not just high MOIC but high IRR. This is not found in industrials or staples, obviously. Anyone who works at a HF but doesn't do TMT is wasting his or her time in the wrong industry. Instead they should be doing industrials or staples levered equities in control private investments or activist strategies. Public markets are best for harvesting dispersion momentum.

Because the absolute return available in those sectors is higher than what's available in others i.e. a hedged tech portfolio will still outperform hedged industrials.

I guess there's more idio vol until the central risk team adds an AI factor?

Incoming Analyst in HF - EquityHedge

This is basically the answer I was looking for. It would make more sense if there’s just inherently more idiosyncratic vol in the space…

I guess it makes sense if cap goods etc are being driven by the cycle so there’s less scope for dispersion

Yes. Think about it this way: an industrial company grows based on rational decision making because companies are governed by costs of capital, inflation, macro factors, and therefore good luck finding stocks that diverge from PMI or industrial production.

On the other hand, consumer and TMT stocks function from the irrational spending on Temu, discretionary purchases for ELF makeup, decision to buy Canada Goose instead of Arcyterx, etc. You can have two very different stories play out. Similarly, everyone hates Fastly and loves Palo Alto. Everyone ditches Adobe and uses Docusign. Everyone loves Cursor AI because it's new and nothing else for coding / agentic AI copilot even existed BEFORE. It's about market share and variance in ROICs. Good luck seeing that happen when you need to spend hundreds of millions of capex over 5 years to even break the barrier to entry or grow 1 pt of market share.... meanwhile my Celsius drink got viral overnight on 0 invested capital and word of mouth free CAC! 

The other response didn't get it and is dumb. It's not about more absolute return. The one word answer to why there is more absolute return is simple: irrationality. The whole point of HF analyst is to make money off of irrational behavior and that is basically the opportunity that every TMT / consumer business exhibits. 
 

How can you tell? Polarization. Go look at how much the price targets or EBITDA estimates 3-5 years out for a stock differ across street. You will find for industrials stocks they are within a 10-15% band. For a consumer or TMT stock you'll find 30-50% bands at times... or the best case where some street believe the company goes bankrupt and is a fraud while others see it as the next 5x. Think about Tesla... clearest example of polarization. If you are right you win big and the other guy who thought it was a bankruptcy blows up. Now do you get that setup for ExxonMobil or a railroad? No. Irrationality. That's why the industrial analyst gets paid $1mm consistently and is better off working in PE and the TMT analyst gets paid $10mm to retire early or gets blown up. But clearly you have better things in life than only $1mm unless you really love covering chemicals and lubricants.

is this a real question? the TMT sector is a bigger sector by market cap and number of companies so it supports more investors. also TMT has more developments relative to industrials so there is more activity to trade on.

Please ask here my friend, I'm also curious.

They’re not buzzwords you moron 

You are the one who lost the message since you’re asking him the same questions he already addressed. Quit sucking off TMT pods for a sec and read what he’s saying, which is basically that it’s not all about capturing dispersion, but rather, strategy-specifics instead

Just try and guess at the broad strokes of his message instead of going through your robot linear-thinking. Not everything is A then B, esp in markets... I'll try to distill his wisdom for you. 
 

@BayesianBets is a seasoned rates PM who understands market nuance - and when its relevant. He saw OP's simple question of:  

"I see bonus info to decide which sector to choose for career. Why TMT pods do better vs other pods when market-neutral (ie low net vol targeting framework)? Is it because they lied about vol target, predictable themes, inefficient flow dynamics? "  (ie can I replicate TMT bonus profits or is # st market driven)
 

Then BB saw the thread's consensus answers and most likely tried to reply to this comment: 

"TMT=high dispersion (outperformers vs laggards) and other sectors like Industrials=opposite, move directionally together, harder to make $$ if market-neutral."

His reply is saying 1. Original question framework is terrible 2. Consensus reply is right on surface, but leaves out valid nuances that changes the answer to the heart of the question.  3. there are higher priority factors you should consider when analyzing alpha from equity space in L/S, and this should be factored in your career decisions. 

• You can't just compare sector returns in low-net vol targeting (and the derived '24 bonuses) to make a decision which is "best" to go into.
• Sector selection itself isn't driving variable for you to generate future PnL - but rather how good your team is exploiting their sector.
• Yes TMT returns were good this year, but the "dispersion" volatility inherent in the space  means you need to be on top of the various flow shifts with drastically higher StDv moves quarter after quarter.
  • 2022 fintech pods blew up. High dispersion =/= outsized returns.
• Consumer Staples had best sharpe with large GMV this year. They probably did this by anticipating how big$ actors (passive/generalist) would react to the lagged emerging fundamental narratives derived off currently known econ/business data.
  • Re-ranking with better framework I'd say Staples outperformed this year compared to other pod groups, and the good ones will continue to do so.
• As a PM/Team/Pod/trader within any sector, the most important thing is to accurately gauge when your sector edge/process is relevant and to size accordingly
• Original question needs to consider sharpe in regards to L/S sector returns.
• Original question needs to consider GMV (gross mkt value $ lev deployed) ""

So I'll re-hash YOUR question reply to BB and I'll try and guess to what he'd say:

(a) are there are more sector specific characteristics to exploit for tmt/consumer than for industrials ?

(b) If not, can you provide an argument for why industrials -- with its various sector specific niche to exploit -- is actually more commercial to cover than tmt?

A. " To give you a linear answer TMT and Discretionary will have more # of characteristics - but you're missing the point. Indus/Staples dynamics are more repeatable and steady where you may only have 5 characteristics (assuming 72% probability of capturing) to exploit any given year, but you can lever up 20 turns and capture a 5% move vs 40 characteristics that are much less predictable (assume 51% probability of capturing ) levering 4x for 25% move. But again the question needs to consider each year these numbers are dynamic and change year to year, q to q and you as the PM need to be on top of that. Also 

B. I think I explained this above and have now lost interest in answering your question. I now see you're a 3rd yr at PE. All the PE guys I know have better reading comprehension than I so I am surprised you couldn't understand BB's reply. 

Associate 3 in PE - LBOs

Ok this is clearly insightful but you lost the message. At the risk of just using buzzwords... you do agree (a) there are more sector specific characteristics to exploit for tmt/consumer than for industrials or (b) it's easier to exploit them for tmt? This reinforces what everyone has been saying. Could you provide an argument for why industrials -- with its various sector specific niche to exploit -- is actually more commercial to cover than tmt?

I can yes.... It's because you've applied the intellectual rigor of a crypto bro and confused sector selection with process excellence, and a fundamental misunderstanding of how skill is measured in MM L/S investing.

We look at alpha in relative terms, not "cross sector momentum". The risk model doesn't allow you to tilt momentum in any specific direction. When investing on a sector basis, you identify long-short spreads by neutralizing industries, countries, and other common factors to isolate skill. The investment process of a team (how they ideate, trade, risk manage & monetize) within that field dictates skill and sustainability in PNL generation. The emphasis is on sustainability to scale a book. Someone wrote “doesn't answer why the TMT guys have been killing it more than others last year.”… it’s irrelevant.

Consider this: PM A might have above-average skill in sizing, making a low dispersion sector more attractive for that team. PM B might excel at ideation but show less skill in sizing & timing, where parts of consumer internet may not lend themselves well to monetization. Others might be top-down thematic traders, good at isolating single names vs. markets rather than pairs, making energy/commodities more suitable.

What you're likely seeing is selection bias due to the "n value" of TMT coverage. Given market size and sector overlap (Industrials covering Semis, Consumer covering Ecomm, Fins covering Payments), you're naturally more exposed to TMT headlines. But you're not adjusting for the massive left tail of TMT L/S folks who don't make money.

I've seen data from two different MMs showing no pattern of TMT teams' PNL being systematically superior on a risk-adjusted basis. If TMT universally offered better risk-adjusted idio alpha under a factor-constrained model, everyone would only run TMT books. They don't.

Fundamental equity analyses is based on process + judgment. PMs don't follow the same investment process, and definitely don’t have the same judgement - if we did, everyone would make money and blow up simultaneously, which we can empirically prove isn't the case. In a high turnover portfolio, you’re increasingly in a zero-sum game - for every winner, there's a loser taking the opposite side. But you don’t see the headlines of the left tail do you?

That said, I'm actually impressed how you've simplified market dynamics so effectively, you've actually transcended into a new dimension of wrong... Saying “it's about market share and variance in ROICs” or “Celsius drink got viral overnight on 0 invested capital and word of mouth free CAC” (speaking of using buzzwords?) you’re analyzing sector alpha with all the statistical rigor of a horoscope.

There’s a reason why MMs have an incredible focus on process + repeatability. So instead of chasing sector myths on CNBC headlines, focus on specific processes and strategy can effectively mine whatever characteristics exist in your coverage universe. That's what drives sustainable alpha generation in L/S. Or don’t… idc :)

Do we though?

I said last year :)