Observations on Scale Economies

We thought that it might be useful if we ran through several themes that Guan Jie and I noticed, and that you might have noticed too. Note that this list is not comprehensive — it merely covers the themes that most leapt out to us as the case writers. If your conclusions differ, that's perfectly fine.

Scale economies often means taking a variable cost and turning it into a fixed cost. We saw this with Ford’s insourcing, and Netflix’s switch to content production (thus turning their variable licensing costs into a fixed cost, amortised over their subscriber base).

We think what’s more important is internalising what this feels like, after looking at so many cases alongside us. It’s trivially easy to talk about what this is, harder to recognise when it’s actually happening (or possible in your domain).

In some rare cases, scale economies lead to a poisoning of the environment for every other non-scale player. We saw this with PayPal. The technical term for this is ‘growth economies’, where the scale player enjoys some advantage due to its position in the market (as opposed to gaining advantages due to its unit costs).

Market makers also have a similar dynamic, where scale players get access to more data, which allows them to build better models, which allows them to hedge their exposure by spreading their risk over a broad range of uninformed market participants, which in turn drives the pursuit of scale, and so on. Meanwhile, non-scale players risk adverse selection.

To quote Kris Abdelmessih:

Competitive equilibrium will mean that the casinos who can bid the highest for the “customer” is the house that can:

a) source the most uncorrelated offsets to the wager


b) has the biggest bankroll

In the trading business, condition A is satisfied by the market makers with the best data/analytics and “see the most flow”. A firm entrenched in both equity markets and futures markets with licenses from both the SEC and CFTC is going to be more efficient at laying off the risks it acquires from serving tourists regardless of the venue they choose to play in.

A and B will create a virtuous loop. The best players will build larger bankrolls which allow them to outbid competitors further which earns them first look at the flow which improves their models and so forth.

Scale economies can work against you if consumer sentiment shifts. We saw this with Ford and Texas Instruments. This was actually quite surprising for both of us — scale economies (at least, in the per unit production cost sense) are really only advantageous if you can take full advantage of the fixed cost investment. Every time there is risk of demand changing, one’s scale advantage may become a burden.

Certain types of scale economies demand some form of centralisation; there are organisational structures where scale does not lead to scale economies. We saw this with Koufu. Again, this is very case specific, but because it shows up in one instance, you shouldn’t be too surprised if you see it elsewhere.

Sometimes you may use pricing to drive learning economies. We saw this with Texas Instruments and Ford. Both cases are notable because a) Texas Instruments was the first to weaponise the learning curve, and b) Henry Ford was (probably) the first person to consciously use price cuts to drive volume, and therefore cost savings.

A variant of the pricing observation is ‘scaled economies shared’. Scale leaders drive costs down and then pass those cost savings to customers. We saw this with Amazon and Ford.

In a variable cost world, scale leads to different advantages. Nike is the odd one out in this sequence of cases. For some structural reason, athletic brands spend the same percentage of sales on marketing, which means that the highest level of brand augmentation goes to the brand that can spend the most.

Scale economies can come from learning economies, from supplier discounts, or from lowering the per-unit cost of production through amortisation. Which case represents which is left as an exercise for the alert reader.

Fung Guan Jie and Cedric Chin, 10 June 2022

Originally published , last updated .

Member Comments