Economic theory is about trying to understand and explain an extremely complex set of processes and one where relationships between actions are not constant, as in the physical sciences, but may change over time. In other words, economic agents may react differently to similar stimuli at different times. That difference may be in terms of degree, in terms of timing or in terms of direction.

And while it might be possible to prove that there is a relationship between two variables, proving that one “causes” the other or is “caused” by it, may turn out to be difficult. Indeed, both may have been caused by a third or missing variable.

A model, quite simply, is a generalisation of what is happening in the real world. By concentrating on the most important variables and making simplifying assumptions about others, a model seeks to describe the underlying process. The robustness of the model will depend crucially on how realistic the simplifying assumptions are.

In this context, the term ceteris paribus is frequently used. This is just a Latin phrase which means other things being equal. For instance, a simple consumption function might be stated as consumption depends on income. Here, other things being equal would mean that variables such as interest rates, wealth, the price level and expectations about the future course of the economy are assumed to be unchanging. In fact, of course, they can and do change and, as they change, they will, to varying degrees, influence the final amount of spending.

Algebraically the simple consumption function would be stated as:

C = a + bY where C is consumption, and Y is income

This equation says that some part of consumption (equivalent to the constant, a) is unaffected by income – there is a certain level of consumption which would occur whatever level income was at – and a further portion which will vary as the level of income varies (defined by the parameter b). C is known as the dependent variable.

At its simplest, a model might be no more than a statement of the relationship between two variables such as consumption and income.

Moving along a step or two, it might entail a number of extra explanatory variables (i.e. those other variables that may cause consumption to change).

Moving further on, it might involve a statistically-based representation of that relationship, using econometric methods (see below) to estimate the strength of the relationship.

If the equation C = a + bY where C is consumption and Y is income, represents the consumption function at its simplest, the Bank of England’s version certainly lies at the other extreme. Click here to see the Consumption Function which is taken from the Bank’s description of its Quarterly Model.

Even more complex, are models which consist of a large number of inter-linked equations. The Bank of England Quarterly Model, for instance, in seeking to explain how the UK economy works, uses over 300 equations. Its aim is to support the Monetary Policy Committee in its monetary policy role of maintaining a stable and low rate of inflation.

Models such as these help to understand how economies work.

But they are also used to create economic forecasts and to answer 'what if' questions. What, for instance, would happen to the economy if there was a sudden depreciation in the exchange rate or a jump in the price of oil? What would be the consequence of various policy changes such as an increase in income tax?

The predictive powers and credibility of most if not all econometric models, whether in the UK or elsewhere, have been put under serious strain in the current economic crisis. Forecasters have consistently underestimated and by a considerable margin the speed and depth of the recession.

Models, we should be reminded, are simplifications of the real world and are based on a range of economic theories about how things work in the real world. Theories such as rational expectations and efficient markets, for instance, which underpin econometric models, have been found wanting and will need to be re-evaluated.

And given the financial sector’s contribution to the downturn, credit markets need to take a more central role in models. As Charles Bean, Deputy Governor for monetary policy at the Bank of England, noted in a speech in August 2009, credit markets should be included in such a way as to allow users “to model shocks originating in the financial sector rather than just as an amplification mechanism”,

Even so, in spite of their apparent short comings, models will continue to be used widely by governments, central banks and companies who wish to read the future. This website includes a smaller model of the UK economy which will allow you to change a number of key macroeconomic variables, and then gauge the effects of those changes against the baseline forecasts. Select Oxford Economics UK Model from the Models and Forecasting menu.

As well as these theoretical models, economists use other types of models in order to understand economic data. An example would be the use of seasonal adjustment models. Economic series are often affected by seasonal factors which obscure underlying economic behaviour. Seasonal models are used to remove those fluctuations.

Seasonal movements arise from a number of factors. They may be based on natural events such as weather changes through the year, imposed events such as the timing of school terms, traditional fixed holidays whether religious or secular (Christmas and bank holidays, for example) or different lengths of months. In addition there are calendar effects such as the changing date of Easter which may not only appear in different months but also in different quarters.

The Bank of England model

The Bank of England model has been designed to model how the UK economy works in practice. In the Bank’s words, the model “describes the behaviour of the economy at a relatively aggregated level that is closely related to the incomes and expenditures recorded in the UK National Accounts. To do this, the model contains formal descriptions of the behaviour of private domestic agents, policymakers and the rest of the world, and their interactions in markets for capital and financial assets, goods and labour.”

For household spending (the consumption function referred to above), the Bank notes that “when deciding on their current level of consumption, and hence their level of saving or borrowing, households are assumed to want to keep their lifetime consumption as smooth as possible. To do this, households can borrow and save using a whole range of financial assets, including domestic equities, corporate debt, government debt, money and foreign assets. In addition, in the short run, households’ level of consumption can be influenced by a variety of other factors, such as short-term fluctuations in their income and their level of confidence about the future.”

Other models

Apart from the Bank of England, there are a number of other large econometric models of the UK economy. These include models estimated and maintained by:

• HM Treasury
• the National Institute of Economic and Social Research

as well as academic establishments such as:

• the London Business School
• Liverpool University

As an example, the National Institute’s model has been in existence since 1969. It has a number of roles to play including its central task of producing quarterly forecasts which appear in the Institute’s quarterly Review. But it also provides background information for debates about macroeconomic policy as well as serving as a focal point for economic research.

The first model designed to capture the workings of a national economy was constructed in the years following World War Two by the Dutch economist, Jan Tinbergen, for the Netherlands and was later extended to the US and the UK. In 1969 Tinbergen shared the Nobel Prize in Economics with Ragnar Frisch “for having developed and applied dynamic models for the analysis of economic processes”.

The first global model, Wharton Econometric Forecasting Associates’ LINK project, was instigated by Lawrence Klein who received the 1980 Nobel Prize in Economics “for the creation of econometric models and the application to the analysis of economic fluctuations and economic policies”.