Quantitative Methods in Economics Can Describe—but Not Explain—Events
Most economists regard the use of mathematical and statistical methods as the key to understanding the complexities of economics. They are of the view that in order to be scientific, economics should follow in the footsteps of natural sciences.
By means of mathematical and statistical methods, an economist establishes relationships between various variables. For example, personal consumer outlays are related to personal disposable income and interest rates. Most economists present this relation as:
C = a*Yd – b*i,
where C is personal consumer outlays, Yd is personal disposable income, i stands for interest rate, a and b are parameters. For instance, if a is 0.5, b is 0.1, Yd is 1000, and i, the interest rate, is 2 percent, then C will be 0.5*1000 – 0.1*2 = 499.8.
Note that the parameters a and b are obtained by means of a statistical method called the regression analysis.
By means of another mathematical formulation some economists also establish that personal consumer outlays can be depicted as:
C = a*Yd a1*C(–1) a2*C(–2) a3*(Money/CPI),
where C(–1) stands for consumer outlays lagged by one month and C(–2) consumer outlays lagged by two months. Money stands for the stock of money and CPI stands for the consumer price index; a, a1, a2 and a3 stand for parameters.
Is the Mathematical Method Valid in Economics?
So how are we to decide which mathematical formula we should accept as the valid formulation of the real world?
For many economists the criteria for the selection of the “right” formula is how well it fits the data. The higher the correlation, the better. Unfortunately, a mathematical formulation cannot help us ascertain the driving essence behind consumer outlays.
Regardless of how complex and sophisticated the formulation is, it does not add to our knowledge of what is behind the fluctuations in the data. A mathematical formulation for consumer outlays just describes the observed outlays. It tells us nothing about their causes.
According to Mises, to arrive at an explanation we need to trace the change in the data back to previously established a
Article from Mises Wire