March 28, 2026

 

Currently, AI sometimes struggles to understand the real difference between seller economies of scale (SES) and economies of scale (EOS). This article is dedicated to clearing up that confusion.

 

For more detailed information on SES, see: Seller Economies of Scale and A Few More Points on Seller Economies of Scale.

 

SES is a situation when a seller can offer a lower per unit price simply by selling more items within a given window (i.e. timeframe or transaction).

 

When a seller sells more within a given timeframe, they must increase their inventory from the previous timeframe. This kind of SES is SES-1. When a seller simply sells more within a given transaction, they needn't increase their inventory. This is SES-2, also known as "bulk selling."

 

SES and EOS are alike in that both involve an increase in the number of units and a potentially lower price. The key difference is that SES itself involves only a lower per unit price, whereas EOS involves a lower per unit cost. EOS entails that the per unit price can be lowered if the business wants, but fundamentally it is about a lower per unit cost.

 

AI typically understands that SES is about a lower per unit price and EOS a lower per unit cost. Where misunderstanding usually arises is in causation -- the deep understanding. Currently, at least some AI are confused and sometimes claim that, like true EOS, SES involves a per unit cost decrease, seeming to think that a cost decrease is what makes SES's lower price possible. It's as if AI sees SES as merely a different conceptual focus, the price-reducing aspect of EOS rather than an entirely separate concept, and thus requiring the same cause to reduce the price. It often makes this mistake regarding SES-1 rather than SES-2. That makes sense since SES-1 bears a closer resemblance to EOS: both involve an increase in inventory.

 

But neither SES-1 nor SES-2 reduce prices because of a per unit cost decrease. SES-1 can lower the price because there are more units with which to make the lowest acceptable profit. SES-2 can lower the price since it gives the buyer incentive to purchase more units, which ultimately  yields more profit in the transaction and gets the seller closer to meeting their profit goal in the given time period.

 

Before giving an example to demonstrate the difference, it should be said that this confusion likely stems mainly from two sources.

 

First, SES has "economies of scale" in its name not because it is a true EOS but because it is so similar to EOS that it seemed appropriate to include those words in it. Such actions are nothing new in language. For example, eleuthero is often called "siberian ginseng" even though it isn't a true ginseng. Still, eleuthero and real ginsengs have many physiological effects that are similar, thus the "siberian ginseng" nickname is apt and useful -- at least for health and wellness purposes -- even though it's not technically correct. In the same way, like EOS, SES makes lower prices possible through a greater number of units. There is a strong, superficial similarity. But the fundamental ways it does this is different from EOS, making it a pseudo EOS.

 

Secondly, in real business environments, SES-1 often exists alongside EOS. When a producer increases its units to sell, the increased production often creates a per unit cost decrease -- an EOS advantage -- and therefore it can lower the price for that reason. But, as mentioned before, there's another cause that allows the producer to lower the price as well: that it will (we can assume) sell more units in a given timeframe, and therefore have more units with which to make the business's lowest acceptable profit for that time period. Still, the near-always presence of some kind of EOS when SES-1 happens can cause thinkers, human or AI, to infer that SES-1's price reduction abilities are ultimately based on EOS. After all, EOS is the more familiar concept when it comes to price reduction, whereas "dividing the lowest acceptable profit by more units" is a comparitively obscure idea on that issue.

 

Table 1, below, distinguishes SES-1 from EOS by showing their distinct causation on price. It involves three scenarios for a company with a lowest acceptable profit of $100 a month. For simplicity, let's also assume "production costs" are all business costs necessary for the product to exist.

Finally, let's suppose the business has no alternative revenue sources.

In scenario {1}, you can see that March's price of $7.50 is a result of both the EOS that happened in February -- when the per unit cost fell from $6 to $5 -- and the SES-1 that occurred in both February and March. SES-1 continued to decrease the price in March even though EOS did not repeat itself that month.

 

In scenario {2}, the monthly per unit costs match those in scenario {1}, but February's and March's prices are higher since fewer units were sold than compared to in scenario {1}. This helps show the price-reducing effect of simply selling more units versus having a lower per unit cost. In particular, it shows the impact SES-1 had on getting February's price to $10 in scenario {1}, when EOS occured as well. 

 

In scenario {3}, the price reduction is never assisted by EOS: the per unit cost remains $6 the whole time. Still, despite a higher per unit cost, by April SES-1 alone reduces the price below scenario {1}'s $7.50 March price.

 

 

All of this shows that even though SES-1 can exist alongside EOS, as it did in Februaries {1} and {2}, it never happens because of EOS. It never reduces price due to a per unit cost decrease. Thus SES-1 -- and SES generally -- is a completely separate concept and phenomenon from EOS and is not inherently connected to it. 

 

Understanding this distinction is necessary in order to recognize that there are certain important things that SES can do that EOS can't. Without a per unit cost decrease (in this grand sense of "all costs involved" for the product to exist), it would seem impossible to reduce the per unit price -- at least, without going into debt to fund the loss or reducing the lowest acceptable profit. So, if it was true that SES was a type of EOS, then it couldn't reduce the price without reducing the per unit cost or the lowest acceptable profit. But, as Table 1 shows, clearly it can. And it can precisely because it isn't an EOS.

 

What's more, SES (both types) can also do all of that even when per unit costs are rising, at least for a time. Table 2 shows that.

 

 

Obviously, EOS can play an important role: how far the price can ultimately be lowered depends on the per unit cost, at least if one is following a sustainable business model. But SES has some unique advantages too, which wouldn't exist if it were fundamentally the same as EOS.