EPI Methodology

The EPI examines the primary segments of an economy – households, firms and government – by incorporating four variables:

• inflation;
• unemployment;
• budget deficits;
• GDP growth into a single indicator.

In contrast to other indicators, EPI does not use complicated predictability maximization procedures but was designed for simplicity, making it easy for the layperson to calculate and simple to apply to the economy.

Raw EPI

The Raw EPI is a very simple metric that assigns equal weights to each of its subcomponents.

The Raw EPI score is calculated as: 100% minus the absolute value of the inflation rate, minus the unemployment rate, minus the budget deficit as a percentage of GDP, plus the percentage change in a real gross domestic product, all deviations from their desired values.

Example, if the inflation rate increases from 2% to 3%, the EPI score falls by 1 percentage point; if an equal change occurs in the opposite direction, the score rises by same amount. Similarly, a 1 percentage point increase in the unemployment rate would lead to a 1 percentage point decrease in the EPI score. On the other hand, a fall in the unemployment rate (i.e. an improvement) improves the EPI score respectively. The same inverse relationship holds for the budget deficit: if the deficit increases, the EPI score falls; if the budget deficit shrinks, the EPI score rises. Finally, if the percentage growth rate of GDP rises, so, too, does the EPI score; when the percentage growth rate drops, EPI falls proportionately.

Weighted EPI

The weighted EPI is designed to overcome problems of consistency during periods of high economic volatility and to make scores comparable across countries, by normalize the data with weights to each sub-component. Weights are determined by calculating the inverse standard deviation of each economic variable multiplied by the average standard deviation of all variables such that the sum of weights is equal to one.
The Weighted EPI is calculated similar to the RAW EPI but incorporated weighted deviation of its components from their desired values.

where W_i is the weight of each component of the indicator, calculated as an inverted ration of variable’s standard deviation to an average standard deviation.
Next, we nominally define the desired values for each of the indicator’s subcomponents as follows:

• the desired inflation rate (I*) is 0.0%;
• the desired unemployment rate (U*) is 4.75%;
• the desired value for government deficit as a share of GDP (Def/GDP*) is 0.0%, consistent with a balanced budget; and
• the desired change in GDP (ΔGDP*) is a healthy real growth rate of 4.75%.

RAW EPI and Weighted EPI

In our research, we calculate both raw and normalized EPI scores. It is worth noting that for developed economies, there are only small differences between the scores. However, for emerging market economies, differences can be significant and normalized data is essential to presenting a true picture of economic performance.

Measuring Economic Performance

In order to make the EPI indicator easier for the general public to understand, we have adopted a simple grading system implementing thresholds close to actual distribution of EPI scores:

Performance	EPI values
A+ (Superior)	Above 100
A (Excellent)	95-100
B (Good)	90-95
C (Fair)	80-90
D (Poor)	60-80
F (Very Poor)	below 60