We, Markets, and GDP

Author: Umang Khetan & Abhishek Gupta, 7th July 2019

GDP is probably the most used metric in news headlines to compare our nation’s prosperity vs. other nations. This is an attempt to understand more about what GDP is and how it matters to different stakeholders such as a common person, and a financial investor.  We attempt to answer it through 4 major questions followed by an annexure on GDP estimation controversy.

Q1. What is GDP and what does it actually tell?

The academic definition of GDP is the monetary sum of goods and services produced (“value created”) by a country in a given time frame, usually a quarter or a year. In more common parlance, it is a measure of the health of a country’s economy, which has strong perceived links to the general financial well-being of the people.

In this piece, we try to bring the academic concept closer home, and draw linkages with the issues a common person faces. We also try to analyse if GDP can help us make better financial decisions. Lastly, an addendum on the controversies surrounding the new GDP series concludes this piece.

Q2. What does the GDP number reveal and hide?

GDP is an estimate. It gives us a sense of how much value various sectors of an economy have created in the past. When policymakers set targets of GDP growth, they essentially try to increase the output of goods and services for achieving a higher standard of living while simultaneously creating the jobs needed to produce those goods and services. GDP therefore serves as a macro “anchor” or goal, to help achieve supporting goals such as employment and industrialization.

When it comes to the level of an individual, the GDP number goes little beyond giving a broad sense of the economy’s performance. Its interesting to explore, if there is any correlation of strong GDP growth being a sign of higher incomes, better jobs, better lifestyle etc.

Unfortunately, tt does not correspond perfectly to the financial well-being of an individual, as we explore in the specific cases below. In fact, the number reveals a lot more about the individual sectors of an economy (details usually glossed over by the common person while looking at only the aggregate number) and a lot less about the economic life of an individual.

Q3. How does a common person see the impact of GDP in his daily life?

-         Does a high GDP growth mean more jobs of all kinds?

-         Does it also translate into faster salary growth and more bonuses?

-         Do prices of goods move in tandem with high GDP growth?

-         Do stock markets move with the same speed as GDP growth?

We tried to use statistics to answer some of these questions. A correlation matrix below is revealing.

Sample of 19 quarters, from Sep 2014 – Mar 2019. Data source: Bloomberg

We realize that correlation is not causation, and that sample makes a lot of difference. Hence, the results are used only for a broad sense.

US employment numbers used to give relatively less reliable employment numbers from India.

Only two correlations are of any significance: wage growth in the US has been strongly correlated to unemployment rate; NSE NIFTY 50 has shown a positive correlation with US nominal GDP growth.

Now let us tackle one question at a time.

(A)  Job creation and job availability

- Conventional wisdom states that high growth translates into more jobs. This is only weakly true, if data is to be believed. This is also highly cyclical: the correlation is stronger in the phase immediately after a recession than in a prolonged recovery period (our sample shows only 25% correlation in the US). After all, haven’t we heard of “jobless growth” in many an economy, and increasingly in India?

- A more granular analysis would reveal strong and weak linkages between the two. White collar jobs are too few to contribute much to the Indian economy, hence the GDP number might indicate very little about availability of such jobs. However, sectors such as construction, that rely heavily on manual labour, show strong correlation between job creation and sectoral GDP growth.

- Sure, faster growth in general can help generate more jobs but this is not guaranteed. There are also issues of productivity and innovation, topics beyond the scope of this write-up.

(B)  Salaries and Income

- GDP is also known as national income. National income does not mean individual income, no matter what proponents of measures such as per capita income would have us believe.

-Fast GDP growth could indicate that producers see fast growth in their top-lines. This could translate into a higher ability of companies to pay more to its employees.

- However, whether the companies will pay more or not depends on the labour supply and demand situation in that specific sector. The 93% correlation between employment and wage growth in the US is very telling: wages depend far more on labour supply/demand dynamics than on the speed with which the country grows (only 27% correlation).

(C)  Prices and GDP growth

-Inflation and GDP growth are not directly linked, generally speaking. However, insofar as GDP growth is driven by increased consumption of goods and services, prices of those items could rise if the supply does not grow fast enough

-More importantly, for an individual, the observed pattern is the shift in consumption from low value everyday items (basic food, clothing) to high value discretionary items (travel, vehicles) as a proportion of total spending, when incomes grow.

(D)  Stock market and GDP

-Lastly, let us look at NIFTY50 performance (LHS) and compare it to real and nominal GDP growth (both LHS)

Quite random, isn’t it? Let us explore this in greater depth below.

Q4. How are GDP growth and financial markets related? Is market cap/GDP a relevant metric to study?

For a layman, a high GDP growth should mean booming financial markets. And hence, you would see “GDP / market cap” ratio being used to make this rationale. Our analysis shows that the ratio isn’t best suited to make financial decisions on account of the following reasons.

-         GDP is an [earnings] metric while market cap is [earnings * valuation] metric, so they are not the same. Also, GDP is a backward-looking metric while market cap being a forward looking metric and the causality is not easily observed

-         Corporate earnings only represent the organised sector while GDP contains all. Hence, this metric is not useful for comparison for an economy like India which is highly unorganised

-         Earnings grow at least partly because of inflation. Therefore, at the minimum, a better metric to compare would be corporate earnings to nominal GDP growth, not real GDP growth

Now, there has been a big debate about corporate earnings not being in line with the nominal GDP growth of our economy, and that they have been subdued for long now. The valuation today reflects the continued expectation for earnings growth to rebound, which has been delayed for long. Now let’s look at those metrics, and corporate earnings decomposition to see what is going on.

Corporate earnings proxy has been taken as Nifty 500 earnings as they represent more than 90% of total listed corporate sector earnings in India.

Corporate earnings (profit after tax) as a % of nominal GDP has fallen to a 15 year low of ~2.5% from a high of 5.5% in 2008 and an average of 4.0% over the last 15 years.

We broke down this data across 3 time periods: 2005-10, 2010-14 and 2014-19 to understand it better.

Now the like to like comparison should be between Nifty ex financials and GDP ex financials as the GDP calculation for the financial sector is calculated basis the deposit and credit growth of the banks sector while NIFTY earnings are not calculated basis that.

While the corporate earnings ex financials exceeded GDP growth in the 2005-10 period, they have been quite subdued in the periods of 2010-14 (7.3% vs. 15.6%) and 2014-19 (6.7% vs. 10.4%).

*Nifty only contains companies that have been present since 2005

Q4A. What could explain this stark difference? Does it mean that we are at a cyclical low and it is time for earnings to catch up to the GDP growth in the economy?

The comparison needs to be taken with a pinch of salt because

-         Composition of NIFTY500 and GDP differs in terms of industry/sector contribution

-         Informal sector and non-listed companies contribute to the GDP, not to NIFTY earnings

-         Large capex across sectors, that contribute to GDP but get lagged in corporate earnings

-         Possible issues in the new GDP series, especially pertaining to manufacturing (addendum)

Q4B. Where are the corporate earnings actually lagging?

If we go deeper into understanding corporate earnings and where have they lagged, financial sector which contributes ~16% of overall earnings, has significantly declined led by the huge NPA built up in the banks’ books with huge provisioning wiping off profits. Even telecom sector has seen huge profit erosion on account of large telecom player JIO entering the sector.

Sectors like IT and Healthcare have also been subdued, with single digit earnings after a strong decade of earnings from 2005-14.

Sectors like consumer stables have been highly resilient which has led to strong valuation for this sector. Sectors like materials and industrials remain cyclical and are very important to time entry into these sectors.

Overview of NIFTY 500 earnings (post tax):

*Excludes Tata Motors for 2014-19 period

**Basis 2015-19 average

Only contains companies that have been present since 2005

If we look at sectors that could possibly present opportunities going forward, they would be financials (revival of the strong franchise PSU banks that have struggled in the past), industrials (focus on manufacturing & beginning of capex cycle), healthcare (revival post regulation disruption and slowing down of US exports), and IT (adoption of new digital business model).

Sectors such as consumer staples and discretionary have seen strong growth which have tapered down recently, with financial sector struggling and could see further deceleration in growth. In fact, the valuations in this space factor continued growth of earnings through strong consumption.

Basis the above research and understanding of the authors, there are a few takeaways that we want the readers to take from this:

1.      GDP is more of a macro anchor than an indicator of individual well-being of the people. Job growth, salaries, and inflation relate more closely to other indicators such as sectoral growth, labour availability, and supply/demand dynamics.

2.      GDP and financial markets are related to the limited extent that earnings growth in a few sectors shall be correlated to the GDP of that sector, with the variation being explained by the extent of informal and formal sector earnings growth, the composition mix difference, capex showing up as lagged earnings etc.

3.      Market cap to GDP is not a relevant metric to look at: market cap is a forward-looking valuation metric and represents only the formalised sector.

4.      Earnings across sectors are widespread and should be studied over a long period to find sectors where odds are in favour of sector being prone to growth and which sector have already experienced a strong period of growth. As a starting point, one could look at sectors where both real and nominal GDP are growing fast, and the contribution to both is being led by companies in the organised space.

The authors have a fascination with understand the macro picture, but also understand the linkages between the macro and micro, given sometimes these correlations shall not be straightforward as seen above. Further, writing motivates us to do more research and reading and, in the process, educate ourselves and the readers too. As always, we highly appreciate your feedback on this and if there are any suggestions on what to write on, do feel free to pass them on.

Addendum on GDP estimation

Controversies around the new GDP series of India (2011-12 base year) are hardly new. Ever since the series was introduced five years ago, experts and laymen alike have questioned the authenticity of numbers on various grounds. The numbers look “too good to be true”, are out of sync with other high frequency data (known as “smell test”), use databases that contain dubious companies’ records, and so forth.

The debate has gained currency again recently after ex-CEA Arvind Subramanian wrote a paper on what he thinks are the flaws with our GDP estimates: he says that the deflators being used especially in the manufacturing sector GDP estimation are “output based” instead of “input based”. To put it simply, the nominal GDP in this sector is getting deflated by a smaller denominator (say CPI instead of WPI), causing the real GDP to look larger than it actually is. His own estimate puts GDP growth at 4.5%!

Gita Gopinath of the IMF, numerous economists in India, and a number of policies think-tanks have also raised questions from time to time on our numbers. One of the common complaints, apart from the one mentioned in previous paragraph, concern the use of Ministry of Corporate Affairs’ (MCA) database on corporate sector performance, which is a departure from the previous series’ RBI database on the subject. MCA database recently came under fire from an audit conducted by NSSO, where a large number of shell companies were found to have contributed to the corporate sector performance, an issue less encountered in RBI’s database which is less timely but evidently more accurate.

Whether it is the issue of deflators or database being used, as users of information the best one can do is to look at alternative sources of performance indicators of the economy, if GDP numbers seem unreliable. RBI has created what is called “Composite Leading Indicator”, which uses high frequency indicators such as currency in circulation, auto sales, electricity generation etc to get a sense of economic performance beyond the use of GDP (this index predates the new GDP series). Many organizations are using curated indices to draw their own conclusion on economic performance.

Authors welcome more discussion on these alternative indicators, but all the same leave it to experts in the field to find out what fixes, if any, are required in the current GDP estimation series.