Actionable insights on equities, fixed-income, macros and personal finance Start 14-Days Free Trial
Actionable investing insights Get Free Trial

Unravelling the GDP Number: The Perils of Arguing About Real GDP Growth


Finance Minister Chidambaram and BJP politician Yashwant Sinha had a public spat recently about GDP growth rates. According to Chidambaram, the statement made by Gujarat chief minister Narendra Modi, that GDP growth during the BJP rule was 8.4%, was plain wrong. In the years that BJP was at the helm, the growth rates were 6.7%, 7.6%, 4.3%, 5.5%, 4.0% and 8.1% respectively, so where was the 8% figure coming from?

Sinha retorted that what was important was what GDP growth they ended with – and when the BJP left the country was growing at 8%, while with the Congress the country is now at 4.8%.

Let’s ignore the political argument for now about who was better. But first, as they accuse each other of fake encounters or terrorism with facts, let us first try and understand what these numbers actually mean.

Is 4.8% bad? America or Western Europe would kill for these growth rates, as would Japan. Is 8.4% good? It doesn’t look like the kind of growth one thinks is great (for instance, would you be overjoyed if your salary was increased by 8.4%? Possibly not.

What does GDP mean?

GDP is the sum of all the value created in a country in a year. Let’s take a simplistic example of a country which has only one factory, which can make 100 apples a year. 100 apples sell for Rs. 1 each, for a sum total of Rs. 100. (This is simplistic, so bear with me). That is the GDP.

Now, the factory can increase the price of the apples to Rs. 1.5 each. That will increase the GDP to Rs. 150, or a 50% increase! Is this good? The answer is no: the same number of apples is being produced, just that they are being sold for a higher price. The higher price is, effectively, inflation.

The “nominal” growth therefore is 50%. So if you subtract inflation of 50% (since the apples went from Rs. 1 to Rs. 1.5) from the GDP Nominal growth, you get 0% “real” growth. In effect, we’ve not produced any more apples, so we don’t have any growth.

Take another scenario. Let’s assume that the factory was expanded to produce 120 apples, whose price goes from Rs. 1 to Rs. 1.1 each. This means the total sales were Rs. 1.1 x 120 = Rs. 132.

GDP grew nominally from Rs. 100 to Rs. 132 (a growth of 32%). But since inflation was 10%, the actual growth is lower, since we only produced 20 more apples (100 to 120), we can only have seen 20% “real” growth. We have taken the nominal growth of 32% and deflated it by the impact of inflation to get a real growth of 20%.

Effectively, the GDP growth number that is bandied about is growth net of inflation; so even if the economy grew by more than 10% a year in nominal terms, in both UPA and BJP regimes, a good portion of this growth was inflation, which has to be “removed” from the GDP number.

Put another way: Would you be happy if your salary was increased 8.4%? If inflation is 10%, your expenses will grow at a high rate than your income, which is untenable. But if you got a 15% increase, then you can say your “real” growth is 5% – since that’s the number you make that is higher than inflation.

It’s all fine when you have just one factory. But it’s next to impossible to accurately estimate GDP when you have a billion people and a land mass the size of India. When you pay an auto rickshaw driver, that payment should be part of GDP, but no one’s counting. When you paid a much higher amount to your neighbourhood vegetable cart vendor than the price in the wholesale market, no one accounted for that inflation.

GDP is fiendishly difficult to accurately know, so there are only different ways to estimate it. The Ministry of Statistics and Program Implementation (MOSPI) has a manual on how they calculate it for India.

MOSPI gathers data for many sectors – Agriculture, Mining, Manufacturing, Trade, and Transport, Banking and so on. This data is accumulated from various sources, and much of this calculation has a large room for error.

For instance, for agriculture, data is gathered for total area cultivated. Only about 85% of this has any degree of accuracy. MOSPI then estimates total value of output from wholesale mandi prices during peak marketing seasons of each commodity. Why should we use mandi prices only for peak marketing seasons (when prices will be low), when some farmers might warehouse and store products for other times? Input information is estimated from sales quantity and wholesale prices of fertilizers, but there again, farmers pay very different prices from what is indicated in the wholesale price index.

This inaccuracy, like the butterfly effect, can be small errors in individual data that, when aggregated will compound into a large number, even hundreds of thousands of crores. That 8.4% could well be 7.5% or 9.5% if you worked figures around.

And then, there is the complication in how you calculate “real” growth. Currently all numbers are quoted in terms of 2004 rupees – that is, assuming we are in 2004 with today’s production, how much more value are we creating today?

The calculation involves deflating the “nominal” number. For agriculture, they deflate both input and output prices. Input prices of fertilizers, for instance, are deflated using the wholesale prices (through the wholesale price index, or WPI) for fertilizers.

Let us assume we sold Rs. 15,000 worth today fertilizer versus Rs. 10,000 worth in 2004. The price in 2004 was 100, and today it’s 120. The nominal increase in sales is 50% (10,000 to 15,000). But the “real” increase, if we attempt to deflate for prices, is 25%

Note: The math is: Compare (10,000 / 100) and (15,000 / 120), which is (100) and (125).

Using the WPI in prices is one mechanism. Others used by the MOSPI for deflating GDP to a real number are “yield per hectare” calculations for agriculture, differences in the Index of Industrial Production (IIP) for manufacturing, Employment data from NSS rounds, CPI numbers (there are four different CPI indexes) and so on.

The degree of inaccuracy in each of these elements is high. Arguing about whether 4.8% or 8.4% means little when, if you were to use a slightly different method, you get a very different answer.

If I were to use the Consumer Price Indexes as an inflation estimate and then look at the GDP numbers, I could easily arrive at this conclusion: In June 2013, we grew “nominally” by 8.14% over Jun 2013. But inflation (consumer prices) was over 9%. So we have negative “real” growth – which means we are in a recession. (Officially, “real” growth is 4.4% though).

And then, even if we wanted to score political points about GDP growth, it would be silly to not consider international events. In 2000 and 2001, the dot-com bust and the terrorist attack in the US caused a recession in the west, but their response of easy liquidity kept money cheap and flowing into emerging economies like India, until 2008 when the dam burst. From then on, money has become even cheaper, and now just the fear that the tap will be shut caused our beloved rupee, and our economy, to hurt severely.

It’s not easy to attribute GDP growth rates to governments and policies. Roads built 10 years ago might only contribute to GDP today. Removing a diesel subsidy will hurt everyone in the short term, but will win laurels after the five year term a government gets. The lack of government control might actually spur an industry (like IT) than others where there are too many licensing issues. In fact, it’s more likely we grow well in spite of our governments, than because of them.


Like our content? Join Capitalmind Premium.

  • Equity, fixed income, macro and personal finance research
  • Model equity and fixed-income portfolios
  • Exclusive apps, tutorials, and member community
Subscribe Now Or start with a free-trial