A $10 rise in oil prices reduces world GDP 0.5 per cent or percentage point ?

Media have devoted a great deal of ink to the impacts of higher oil on economy based on the findings of a study done, for example, by the International Energy Agency (Analysis of the impact of high oil prices on the global economy, May 2004, available at its website.) Here are some examples how this is reported in media.

“According to the agency’s report, any sustained $10 increase in the price of oil would increase unemployment, inflation and budget deficit across the world and cost the global economy at least $255 billion within one year…In Asia, the energy agency estimates, the effect of a $10 increase in oil prices would result at least in a 0.8 percentage point reduction of growth.”
International Herald Tribune, (return of an economic demon: oil shock), 4 May 2004.

“World GDP growth may have been at least half a percentage point higher in the last two or three years had prices remained at mid-2001 levels.”
The New York Times, (higher oil prices are damaging global economy, a study shows), 4 May 2004.

“A $10 rise in oil prices subtracts at least 0.5 per cent from world GDP.“
Financial Times, (high oil prices hitting world economic recovery, IEA warns), 4 May 2004.

“Loss of GDP would average 0.8 percent in Asia.”
Reuters, 4 May 2004

At first glance, two observations can be made. First, big number gets an immediate attraction as said by Joseph Stalin: “One death is a tragedy, a million deaths is a statistics.”

But the most important thing to observe is that while the New York Times says “percentage point” Financial Times says “per cent.” The assertions that if oil prices remain at today’s level economic growth will be reduced by X percent or by X percentage point do not have the same meaning. They are not synonyms.

Reporters are not alone in this misuse. They are misused even in the original report. For example, Table 2 of the IEA study gives the impact in percentage points but in the text the same numbers are used in terms of percent. And media repeated that.

In the original study it says, “Philippines would lose 1.6% of its GDP in the year following the increase, and in India 1%. China’s GDP would drop 0.8%...” But then in Table 2 the numbers are given as deviation from base case, in percentage points. Asia: -0.8; Philippines: -1.6; India: -1.0; China: -0.8.

Percent refers to a relative change (some fraction of an original amount) rather than an absolute change.

Percentage change is a difference divided by some base number.

The term percentage point is used to get around an ambiguity in English when comparing two different percentages. It is a simple addition or subtraction of two percentages.

In other words, it is the number that a percentage increased or decreased - not X% of something. For instance, if a bank increases the interest rate from 10 percent to 11 percent, the percentage of change is an increase of 10 percent, but the interest increased by 1 percentage points, from 10 percent to 11 percent.

What does that mean? If GDP growth rate is 4.5 per cent in Philippines, a -1.6 percentage point decrease means that the growth rate for GDP would be 2.9 per cent, which is alarming. But if the drop is -1.6 per cent, then GDP growth rate would be 4.428 per cent.

So, who is guilty? The interchangeable use of these two notions even in the original report is definitely innocent but the journalists simply blindly repeated it. And that shows where journalism stand today!