3.2 Applying scientific methods and reasoning
Scientific reasoning involves constructing theories to explain how the world works. In general, the simplest explanation is preferred. As new information about the world becomes available which throws doubt on a theory, then the theory is modified. Consider the following example of how the process can be applied to an economic problem.
- We have the following information about the world - supermarkets are selling more organic apples than before - and we want to know why this is the case.
- Now we construct a theory - the volume of organic apple sales depends upon the price at which they are sold - this is a positive statement.
- Next we state a hypothesis - the volume of organic apples that a supermarket will sell rises if it reduces the price of organic apples - another positive statement.
- Then we test it by looking for evidence to support the price-sales link in different countries and for different time periods.
- If the evidence supports the hypothesis
- we accept it, and
- move on to more refined hypotheses, such as those relating to how big the response by consumers is to price changes.
- If the evidence does not support the hypothesis, we either
- modify the hypothesis, or
- reject it.
This process of trial and error gradually leads to a theory that fits well with experience. It enables the economist to
- explain what has happened
- say what might happen in the future when prices change
Can the theory be tested?
In the organic apple sales example, yes.
But what happens if the hypothesis is not testable?
You have to try to frame ideas of how the world works in such a way that evidence can be found to support or reject the idea. Incidentally, this limits what we can theorise about. In practice it is sometimes possible to find a way round the unmeasurable by using a proxy variable. This is a variable that reflects the characteristics of the other, unmeasurable, one. So in the case of placing a monetary measure on the value of the environment we can
- look at the price paid for related goods
- ask individuals using survey methods what value they place on the environment
This is not perfect, but proxy variables can be a way round such problems.
Why do economists make assumptions?
In everyday life assumptions are made because nothing would ever be achieved without them. Whenever we open an economics textbook we do so with the underlying assumption that, for whatever reason, economics is a subject worth studying. We don't examine this assumption in detail every time we open an economics book (unless we are extremely disillusioned) because otherwise we would make very little progress with our reading. We also learn from experience which assumptions have lasting validity.
This illustrates an important reason why economists make assumptions. Assumptions save time and concentrate the mind on the problem to hand, and when well-proven become part of the body of established knowledge.
The use of assumptions in economics also relates to the need for an analytical method that provides clarity, scientific rigour, and flexibility.
Consider the theory represented by the positive statement that we made earlier: 'the volume of organic apples that a supermarket will sell rises if it reduces the price of organic apples'. The theory is stated with the assumptions that: (a) the quality of apples remains constant; (b) consumers are made aware of the price fall; and (c) consumers prefer more apples to fewer apples.
The use of these assumptions helps clarify the relationship between organic apple sales and organic apple prices. It does so in a scientific manner by explicitly recognising that organic apple sales can also be affected by other variables such as quality, price information, and consumer preferences. In a controlled laboratory experiment these variables would be kept constant. In economics it is not possible to conduct laboratory experiments. Instead, economics depends greatly on the techniques of statistics for testing its theories and hypotheses. These techniques can be used to identify the relationship between different variables, such as sales and price.
Setting out underlying assumptions when developing and testing a theory or hypothesis also provides the analyst with the flexibility to vary the assumptions at a later date. Modifying assumptions and examining the effects of these modifications on the original theory provides a methodical way of examining economic problems and the variables that link cause and effect. For example, once we have established the nature of the link between apple prices and apple sales we can move on to look at the link between sales and quality.
What is an economic paradigm?
It has been observed that research tends to cluster around 'big ideas' called research paradigms. Once a particular paradigm is accepted as dominant, people often research issues adopting these central tenets. In the 1950s and 1960s the dominant economic paradigm regarded markets as being prone to failure, and the duty of government was to intervene and correct these failures in an impartial manner. People became disenchanted with this approach as a result of the 1967-1977 decade of inflation and unemployment, leading to the ascendancy of a new paradigm - one which regarded government intervention as being more likely to worsen than improve market failures. In this paradigm, the job of government is limited to the role of ensuring free competition and making companies respond to shareholders. In recent years, this paradigm has begun to be displaced by one that re-emphasises the role of government in correcting market failures. This is especially the case in environmental economics, although the 'credit-crunch' of 2008 drew attention to problems in financial markets that were possibly under regulated.