Economies of scale

This note is primarily directed to students. It is the first in a series on strategic cost analysis. One of the most important cost drivers is economies of scale . A key point to note is that diseconomies of scale, which textbook authors like to describe, hardly exist in the real world.

Definition. Economies of scale refer to the unit cost being lower when a larger volume is being produced. There are a couple of reasons behind economies of scale:

• Large amount of fixed costs. Fixed costs are not varying depending on the volume of products produced (e.g. image advertising, research and development, capital investment, general and administrative costs). If a company produces a larger volume, then the average per unit cost is reduced. All companies have fixed costs, but unless the fixed costs are larger than ~20% of total costs, economies of scale are not important. This is the most important factor by far.
• Large companies can negotiate harder with their suppliers. This is traditionally not discussed as economies of scale, but in some industries the large companies can negotiate large discounts from their suppliers. The effect would normally result in a lower variable cost.
• The cube-square rule and its variants. The volume of a cube is proportional to the cube of the length of the side. However, the material cost of the tube is proportional to the square of the length of the side. A larger power plant or refinery will have lower average total cost as a consequence. Conceptually, a similar mechanism is at work for a company that carries a lot of inventory. 

A related term to is minimally efficient size (MES). For practical purposes there will be a volume of production that will result in a fairly efficient operation. If a company produces more than the MES, there will only be minor cost savings. To understand how the MES is affected by fixed cost have a look at figure 1. An industry with 20% fixed costs is compared to an industry with 80% fixed costs. With 20% fixed costs, economies of scale are not very important and the MES is around 80. With 80% fixed costs, economies of scale matter greatly and the MES is around 4-500. (There is no mathematical was to exactly determine the MES. A judgement needs to be made.)

Figure 1. Illustration of economies of scale.

Diseconomies of scale do not matter. In introductory microeconomics, it is often argued that above a certain volume of production, there are diseconomies of scale. However, empirical microeconomics (i.e. industrial organisation research) have concluded that such costs are rarely occurring in reality. Diseconomies can exist in the short run (< 1 year). If a factory is already at full capacity it is possible to run the factory 24 hours per day as well as increase the speed of the production line. Hiring shift workers or paying overtime will cost extra. Running the machines at higher than the recommended speed will result in more breakages and repair costs. However, in the longer run, companies can increase the capacity by hiring more staff and by building an extension to the factory. In strategic management, we are mostly concerned with the long run costs (i.e. more than one year ahead). Diseconomies of scale are rarely observed for long run costs.

If diseconomies of scale were to exist, two large companies merging would see their total average cost per unit increase. Such data are never observed. The most likely outcome when two large companies merge is a small decrease in the total average cost per unit. During a two-three year period the merged company might actually have a higher cost per unit due to integration problems. In contrast when two smaller companies merge, a larger decrease in the total average cost per unit can occur. 

So why do the microeconomists persist in talking about diseconomies of scale? There are two reasons: 

• Microeconomics is largely a theoretical field and empirical research is rare. Empirical microeconomics is actually a totally separate field called industrial organisation. (Not to be confused with industrial organisation research in psychology or management). These empirical researches normally do not bother with the research done in microeconomics!
• Microeconomics needs a mechanism to explain why there is not a monopoly in every industry. If the average total cost curve would continue to go down with increased volume, then the end result would be a monopoly producing at the lowest cost. Since this is not observed, the microeconomics have "invented" diseconomies of scale. (However, this is not necessary in real life. First, a flat portion of the economies of scale curve will also function as a barrier to getting bigger. Second, most products are differentiated to some degree. Varied buyer preferences also leads to most industries not being a monopoly )

(Unimportant side comment: The term minimally efficient size originates from research looking at short term diseconomies of scale in manufacturing. I use it to indicate an estimated size, at which most economies of scale are already reached.)

Data sources. Annual reports are a good source for determining the various types of costs occurring in an industry (cost of goods sold, sales cost, capital investment, labour, R&D, etc.). However, additional information about the value chain in particular is required before costs can be categorised as either fixed or variable.

Implications. If the proportion of fixed costs is substantial, the cost position of the major companies needs to be assessed in detail; in particular whether they are located above or below the MES. If the proportion of fixed costs is smaller, economies of scale is going to be less important.

There are implications for a few of the forces in an industry analysis. The comments are general tendencies and other unrelated factors will also influence the strength of the forces.

• Threat of new entry. Economies of scale matters to the extent that it is costly to reach the MES. If the MES can be reached with a modest capital investment, the threat of new entry is higher. (This is especially true if the investment can be reversible. Not normally the case, but possible.) If the MES cannot be reached with a modest capital investment, the treat of new entry is reduced.
• Rivalry. If economies of scale is present, there is potential for different firms to have different cost positions. Different situations can occurt: (1) If there is a clear market leader in the industry, it is likely that the other companies will defer to the market leader. In such situations the market leader will set the price level so that all companies in the industry will survive, in effect it will be the least important company that (indirectly) determines this price level. (2) If there is no clear market leader (or if the other companies refuse to defer to the market leader), the situation is more complicated. (2a) If all companies are above the MES, then economies of scale do not matter greatly. (2b) If on the other had some companies are below the MES and some above, then it is possible that there will be price competition. Some companies would try to lower their prices to gain market share and thus achieve a lower cost position.

There will also be consequences for strategy analysis. The choice of generic strategy determines how important economies of scale are for an individual company. Let us look at the different strategies in an industry with fairly high economies of scale:

• Low cost strategy. The low-cost position can be gained by focusing on economies of scale. It should be noted that there are additional ways to achieve a low cost position, e.g. economies of capacity, novel configuration of value chain. If a company were to focus solely on economies of scale, it would be unlikely to reach the low-cost position.
• Differentiation strategy. Economies of scale are less important as long as the company is producing above the MES. The company's products add other perceived benefits that compensates for a higher cost. However, if the company is producing below the MES, it is severely disadvantaged and will not be very successful.
• Niche strategy. It will be hard for the company to have a low cost position, but that is often not needed as long as the needs of the selected customer segment are addressed. It could be prudent for the company to be above MES, but that is often not possible.