4 Specific Knowledge

We started the discussion of economic theories with transaction cost economics in the previous chapters. The emphasis of transaction cost economics is on the value that organizations create by coordinating different functions in different departments. Transaction cost economics of hierarchical organisations has a top down flavour that focuses on the coordination role of top management. In this chapter, I introduce the notion of specific knowledge because it will help us think about when top management should not take the decisions. That is, I will answer the question when decisions should be decentralised to the departments in the organisation. Decentralisation leads to new measurement and control problems which can be solved or at least alleviated by strategic management accounting tools. In one of the following chapters, I will expand the notion of specific knowledge to the broader category of human capital where investments in human capital are an alternative mechanism for organisations to create a difficult to imitate competitive advantage.

General and Specific Knowledge

This chapter focuses on the problem of assuring that the person or department with appropriate knowledge decides how to implement a strategy, which projects to fund, or how to execute them. I make a distinction between two types of knowledge: general knowledge and specific knowledge. The distinguishing feature of specific knowledge is that it is difficult to communicate (Jensen & Meckling 1995). If a person has specific knowledge that means that they cannot just explain their insight to other people. For instance, the design sensibilities of Jonathan Ive or Steve Jobs at Apple and the experience and negotiation skills of Warren Buffett are more difficult to transfer to other employees. Production workers, such as the manufacturing employees at Toyota, gain unique insights in their job just by doing it. It is costly and often very difficult to impart this knowledge to new employees. They will only learn through their own experience.

In contrast, the market price for a product or service is the quintessential example of general knowledge. If sellers’ cost of production goes up, the market price will increase. If the final customers value the buyers’ products more, the intermediate market price will increase. In a well functioning market, the market price contains all the information that buyers and sellers need about the product. If there is more demand for the product (by the buyer), the price will go up. If there is more supply of the product (by the sellers), the price will go down. The market price is known to everyone in the market and can be understood by everyone in the market and is therefore general knowledge.

There are multiple reasons why information is difficult to communicate. I will focus on two. The first reason is that we humans have cognitive limitations. We cannot digest and remember an infinite amount of information. That means that in all but the smallest organisations, it is unlikely that one person will have all the knowledge necessary for the organisation to be run perfectly. In addition, when we learn new knowledge, it takes time. Just reading this textbook is not enough for you to have the same knowledge that I have ^[1]. When we learn a new skill or new knowledge, we need to practice which takes time. That means that we are learning new knowledge that is difficult to communicate.

The second reason why specific knowledge exists is that a lot of knowledge is specific to an environment and the people who work in that environment know best what is important. For instance, sales people interact a lot with the firm’s customers and as a consequence they know a lot about the preferences of the customers. Similarly, the production team knows the production process the best. They will be able to identify possible improvements in the production process. If a department needs to transfer this information to the headquarters some of the knowledge will get lost in translation. Every report or presentation by the production managers will necessarily leave out some details which might be important for the decision about the production process. The fact that the production managers have to write a report or must have multiple long meetings with headquarters is also a sign that it is costly to transfer their knowledge.

Markets and Specific Knowledge

As we have seen before in the story of the mine, there are two possible solutions to assure that the people with knowledge take the decisions. The first solution works in the well functioning market. A market operates best when the rights to assets are in private ownership and these rights can be bought and sold. This means that the owners of the rights transfer the rights to somebody else and the new owners reap the profits of using the machine. If you have the ownership rights to a machine, you can sell this machine and you get the sales price when you sell. You can also use the machine to produce goods and sell the products to consumers. In such a case, market forces will assure that the owners of the rights will sell them to the most knowledgeable people in the market. These people know what to do best with the rights and can use the machine most efficiently. The machine is more valuable for knowledgeable people and they are willing to pay the highest price for the rights to the machine. Because that price will be higher than what the current owners can earn from the machine, the owners will be happy to sell.

Organisations and Specific Knowledge

The market solution does not work in organisations and it is obvious why. The sales department can not sell the goods to consumers and keep the money. The price that consumers pay for the companies products belongs to the company. Similarly, the production department can use the organisation’s machines but it is rare that they are allowed to sell the machines. And even if they are allowed to sell the machine, the proceeds from those sales do not belong to the production department but to the organisation. In these examples, the sales person and the production manager do not directly benefit from the decisions that they take and this sets up a potential conflict of interest.

Numerical Example of an Organisation

We can use a simple numerical example to illustrate the problem. Let us assume that there is a manufacturer of computers with two departments: a production department and a sales department. The production department can invest in a technology to produce shiny white computers which can be sold at a higher price to consumers than regular grey computers. The production department is evaluated as a profit centre with a fixed transfer price for each computer that it manufactures and transfers to the sales department. As a result, the production department does not profit from an investment in the production technology for white computers. Whether they deliver a white computer or a grey computer does not matter for the transfer price in our example. The sales department can invest in more attractive lighting in the stores which will increase customer traffic and thus customer demand. The investment is so substantial that the sales department alone will not profit enough to invest ^[2].

To make this example more tangible, I will put some numbers on these investments. We have four possible options: the production department can either invest or not in white computer technology and the sales department can either invest or not in better lighting in the stores. In Table 4.1, you can find the profit of the production department for each of the four options. The key insight is that the production department would rather not invest independent of what the sales department does. If the sales department invests and the production department does not invest, the production department has a profit of 450 while investing gives a profit of 350. Similarly , when the sales department does not invest and the production department does not invest, it has a profit of 300 while investing gives a profit of 200.

Table 4.2 shows the profit for the sales department with the same four options. The sales department is in a very similar situation. No matter whether the production department invests or not, the sales department is better off not investing in the lighting for the stores. The profit of not investing is always higher than the profit of investing.

In our simple story, we can just add up the profit of the production department and the sales department to get the profit of the firm. Table 4.3 shows that the firm has the highest profit when the production department and the sales department both make the investment.

This situation is the problem of asset specific investments in transaction cost economics. The sales and production department are better off not to invest but the firm as whole would benefit from the investments ^[3]. This is a good place to think about what the solution is to this problem. You can even go back to the chapter on transaction cost economics because coordinating asset specific investments is exactly the problem that transaction cost economics identifies as the key problem in economic governance.

The solution is simple because the story is simple. The best option for the firm is for top management to take the investment decisions and let the departments execute the strategy that is implied by these decisions. This is why transaction cost economics explains the benefits of the hierarchical organisation where top management coordinates the investments.

Extension with Uncertainty and Specific Knowledge

In the simple story above, I assumed that top management exactly knows the benefits and costs of both investments. This assumption is not always useful. It is not hard to imagine a situation where the production department has specific knowledge about the true cost of a new production technology and the sales department has a better understanding of customer demand and how customers will react to changes to the stores. For simplicity, I will assume that the true costs and benefits are given in Table 4.1, Table 4.2, and Table 4.3. The specific knowledge of the production and sales department means that they know respectively that Table 4.1 and Table 4.2 are true and therefore together they know that Table 4.3 is true. However, because the tables are based on specific knowledge they cannot be communicated to top management.

Let us further assume that because top management has less understanding of the investment decision, they believe that Table 4.3 is a possible outcome but they are not sure that they can avoid the worst case scenario in Table 4.4 where the shiny white computer production technology is more expensive than expected and the demand is lower than expected with the introduction of better lighting in the stores. Under the worst case scenario, the firm would be better off not to invest in either technology.

This minor change to the original story complicates how the firm should make investment decisions considerably. It is no longer guaranteed that top management will take the decision that benefits the firm as a whole because the necessary knowledge to make the decision is in the departments. Again, this is a good place to think about how top management can solve this problem.

There are two broad categories of solutions. The first one is a top-down approach. Top management can try to make the knowledge in the departments readily available to them by requiring the departments to quantify their specific knowledge. For instance, the production department can use a cost accounting system to quantify the cost of producing the white computers and the grey computers. Similarly, the sales department can use a customer profitability analysis to understand the willingness to pay for grey and white computers in different customer segments. Quantifying specific knowledge is one of the key contributions that strategic management accounting has to better strategic decision making.

The second solution is a bottom-up approach. Top management can decentralise the decision making process to the departments which immediately introduces a new problem. The departments will not take the best decision for the firm as a whole, even in our most simple story, and thus, top management will have to design incentives to make sure that departments make investments that are in the best interest of the firm as a whole. Importantly, top management does not know what the optimal decision is a priori. If they did, they would just take the decision themselves and they would not need to bother with decentralising decision making.

In our story, a straight forward solution is to tie the compensation of the departments to the firm’s profit. There are potential problems with this approach because factors beyond the control of the departments could affect the profit. For instance, the production department needs to trust that the sales department will make the right investments and vice versa. The production department also need to trust that customer demand will not be affected by macro economic factors or decisions by competitors.

The problem of designing incentives that allow the firm to decentralise the investment decision to the departments is the subject of a future chapter on agency theory. In this chapter, I will explain the properties of good performance measures to motivate the departments to make decisions that increase total firm profit. Here, I want to highlight that management accounting tools play an important role in defining the scope of decentralisation and the incentives for the departments. If the department has a budget to invest in new technologies, the firm has granted them some decision rights. If a department is evaluated at the end of the year on how much money they spend of the budget, they will be incentivised not to overspend. Budgets are at the heart of many investment decision rights in hierarchical organisations and therefore they will be the first strategic management accounting tool I will discuss in the second part of this textbook. However, the other management accounting tools such as cost accounting and balanced scorecards play an important role in measuring the performance of departments and investments and thus, they are also used to design better incentives for decentralisation.

Summary

In this chapter, I introduced the notion of specific knowledge as knowledge that a person or a group of people has and that they cannot easily or quickly communicate to others. In combination with transaction cost economics, it gives us a powerful framework to understand hierarchical organisations and the need for management accounting tools. The transaction cost framework emphasises the role of top-down process because of the role of top management to coordinate the different investments in the organisation. Specific knowledge emphasises the limitations to the top-down process because top management will not have all the necessary specific knowledge. As a result, top management needs to use management accounting tools to translate, collect, and communicate specific knowledge to top management or they need to design a more bottom-up structure where decisions are decentralised. The latter solution also requires management accounting tools to measure and evaluate the performance of the decision makers. All these management accounting tools have transaction costs associated with them. The role of top management is to find the balance between good decision making and transaction costs.

References

Jensen, M. C., & Meckling, W. H. (1995). Specific and general knowledge and organizational structure. SSRN Electronic Journal, 251–274.