**The base effect is the one that occurs when we choose different bases to compare temporal or cross-sectional data, resulting in different values, and therefore, they can be interpreted incorrectly.**

Simply put, this effect occurs when we decide how we are going to compare data. The choice of a reference base is essential, since it allows the data to be compared with reasoned logic. If we choose wrong, we will most likely make a mistake when comparing.

An example would be when the price of a product that is $100 drops to $50 over a period of time. A priori we can say that it has been reduced by 50% because we take as a basis to compare those 100 dollars of the increase. The calculation is as follows:

[(50/100) -1] = 50%.

Now, suppose that the price returns to its initial place. That is, it goes from $50 to $100. We could mistakenly think that the rise is 50% just like the fall. But in this case, since the base we use to calculate the fall is the starting point of $50, we can verify that it has increased by 100%, from $50 to $100. Since the base this time is 50 dollars. The calculation is as follows:

[(100/50) -1] = 100%

Therefore, we see that the increase has been greater than the decrease when comparing using different bases.

## Inflation, GDP and Base Changes

Inflation is a continuous rise in prices, measured by a parameter such as the Consumer Price Index (CPI). Normally, the annual is compared with its equivalent of the previous year and the same happens with the monthly, using the variation rates to do so.

This variable is highly influenced by the base effect. It is not the same to compare when the base value was abnormally high or low as if it was more or less moderate. The same thing happens if instead of comparing with the previous year we do it over two years or more.

For its part, the nominal GDP of a country is an indicator of economic growth, which should not be confused with development. Therefore, it indicates the variation in the production of goods and services in a period, valued at the prices of that period.

The GDP has a clear relationship with inflation, since it affects it by increasing its value and distorting the interpretation. To avoid this, real GDP is calculated, which does not take into account the effect of prices because it uses a fixed base (a given year) as a reference.

## The Basis Effect on Investments

In times of crisis there is a tendency for central banks to carry out expansionary policies which, in turn, will fuel inflation. These measures are taken precisely to avoid the opposite effect or deflation.

This phenomenon is known as *reflection* (reflation), coined by Irving Fisher in 1929. Well, there is another term related to the previous one *reflaction trade* and it tells us in which financial assets it is convenient to invest with inflationary pressures.

Thus, the tendency is to do so in equities and not in fixed income, above all because these expansionary policies lower interest rates. In this way, the dividends of the companies can provide higher returns to the investor.

An example would be the COVID-19 pandemic and the global lockdown carried out in 2020. The expansionary policies of the different central banks in 2021, to get out of the recession, caused inflationary pressures. In anticipation, investors took refuge in equities.

## Base effect example

Finally, let’s see an example of this statistical effect with a fictitious country. The data, by way of example and exaggerated to see the effect clearly, are nominal GDP, real GDP, the CPI and the variation rates of the first two.

If we look at the image, we see that nominal GDP grew by 50% in year 2 and 100% in year 3, taking into account the price increases shown by the CPI. Discounting these in real GDP, we see that it was not a big deal, with the rise being slightly more than 40% and 75% respectively.

The difference in this example with extreme data may not seem like much, but let’s imagine the nominal GDP of a country like Spain, which is just over a trillion euros. We see that the base effect is more than relevant when comparing data.