Rajan Govil wrote a good piece on India’s growth rate rebound to 6.3% (y/y) in the fiscal second quarter of 2017-18 (July to September). Rajan raises three issues with the growth statistic:
(1) Bank credit growth and GDP growth
(2) IIP growth and growth in Gross Value Added (GVA) manufacturing
(3) Inferring GVA manufacturing from corporate data
I think these are important questions. If tax collections suggest vigorous economic activity, then at least a majority of indicators should point in a similar direction.
For instance, the output of core industries does not exactly offer reassurance that GDP growth could be of the order of magnitude reported. [Link]
GVA Manufacturing and IIP can and should be closely related, if measured correctly. Also, Rajan questions the use of data from corporate conglomerates to infer manufacturing GVA. That is a valid question too.
Another friend – who would prefer to remain anonymous, I guess – responded to Rajan’s critiques. His substantive points are:
I strongly believe India’s industrial production figures are useless since it adopts a fixed factory approach for most sub sectors (i.e. fixing the factory from which data is collected), and not a flexible capacity/sector-based approach. This methodology is flawed for two reasons. First, if the capacity expands by the introduction of new competitors, it should Be counted as extra production, but it happens only in some sectors, such as Electricity production. Second, if a company shuts down or stops production, it shows activity falling sharply. This happened when Nokia shut down its Chennai factory, leading to a collapse in telecom equipment IP, at a time when India was probably selling more than 50mn phones a year, and is happening recently with Havells shutting production in one of the sample factories, which led to the fittings IP falling dramatically, at a time when LED bulb production in India has grown probably by millions in a short period.
The Index of Industrial Production does not also distinguish between a Mercedes and nano, which the GDP does. Indeed the value addition is different. That is precisely the reason why looking at financial statements of corporates by sectors is more useful, since it does actively track what the corporate sector is producing and profits it is generating, rather than by an old school soviet way of measuring heads.
To link bank credit growth as a way of financing economy is a very old school idea. If one splices up banks into public and private, its public sector which has shrunk massively, private sector lending continues. Second, with the renewal and resolution of several projects and the government moving to an EPC based infra model rather than PPP, private sector demand for credit has dried up, but it does not mean activity has slowed down. Further, government agencies are actively tapping the corporate bond market, and I think it is time RBI needs to consider a total social financing indicator a la China for India.
Bank lending to industries is not happening at all. It is contracting. Private sector banks lend to consumers. That is why personal loans are advancing at a double-digit rate. Yes, there are non-banking sources of financing. Over the last five years, they have always been there in India. There is no evidence that they have taken up the slack in bank credit growth. That said, you are absolutely right about the need for a total measure of financing, and not just bank credit data. See here and here.
The RBI Annual Report for 2016-17 has a Table (Table II.6) on ‘Total Social Financing’ (TSF) for India [Link]. It is not called as such, though.
As you can see from the table above, total financing fell in 2016-17 and the flow has been negative in the fiscal first quarter of 2017-18. Non-banking sources have not been able to offset the banking credit deceleration.
My biggest reason to remain suspicious of the GDP data (new series) is that it revised up the growth rates for 2012-13 and 2013-14 when things were quite bad and the government was massively shrinking the current account deficit. There is absolutely no reason to think that real GDP growth was 5.5% and 6.4% respectively in those two years. I had relied on many sources of other real economic data to show that the GDP growth for those two years appears overstated considerably. Since it is the same series that we are working with now, my scepticism remains.
Incidentally, Sajjid Chinoy of JP Morgan has also mentioned, in passing, the disconnect between the growth rate of GVA in Manufacturing and the growth rate in industrial production.
Finally, one thing that no one seems to have mentioned is the decline in the share of Gross Fixed Capital Formation in the overall GDP. In constant prices, it came down to 28.9% from 31.0% in 1Q2016-17. In nominal terms, it was down to 26.4% from 29.2% in 1Q2016-17. Government Final Consumption Expenditure as a share of GDP has risen in this period.
Not the stuff of sustained high growth rates.
Postscript: For the sake of completeness, I must mention this interview of TCA Anant, India’s Chief Statistician, by THE HINDU. It is useful and thoughtful. His point on why the fiscal deficit had already hit 96% of the projected annual deficit for 2017-18 is obvious, in hindsight, but many (including Yours Truly) did not think about it:
A lot of people have picked on the CGA report saying that the fiscal deficit is 96% of the total. They have, as usual, jumped to all sorts of conclusions which in my judgment are wrong. This is something you should have expected. Why? What did the government do this year? It preponed the Budget calendar to allow government expenditure to start from April 1. There is enough evidence to suggest that that did happen. The Q1 government expenditure compared to last year was much better.
Therefore you would expect that by the end of Q2, the average government expenditure level would be higher than what it was last year. Many expenditure management committees have pointed out that the earlier tendency of delayed bunching expenditure in the last quarter is very bad for both the quality of expenditure and fiscal management.
There were a number of recommendations about how the government should better manage its expenditure so as to minimise the amount of expenditure that takes place in the last quarter and last month.
One consequence of this is that, during the year, the fiscal deficit is going to rise because the revenue profile has not changed due to this manipulation of budget dates. The government has made some efforts to push the revenue profile back by changing the advance tax rules. Some of that was done this year too, but those effects will be small. By and large, the revenue profile would remain the same as last year but the expenditure profile has changed, so the logical implication is that the fiscal deficit will rise at this stage. The surprise it generated befuddled me.
The question is, will the government meet its revenue targets. Nobody has made any analysis so far to suggest that it won’t.