Author: Pramod Sinha

Pramod Sinha is a researcher at the National Institute for Public Finance and Policy.

• Measurement of exports in India, 15 May 2017
• How to identify manufacturing companies for GDP estimation, 4 October 2016
• The measurement of Indian manufacturing GDP: problems and some solutions, 30 August 2016
• How bad is IIP growth after controlling for Diwali effects?, 13 January 2016
• A better output proxy for the Indian economy, 21 July 2013

Measurement of exports in India

by Radhika Pandey, Ajay Shah and Pramod Sinha.

The Indian statistical system has many difficulties. In general, a certain mistrust of official statistics is well placed. Every user of data must skeptically examine all data that is sought to be used. In this article, we kick the tyres of exports data, juxtaposing two different sources. We are pleased to report that there is no large discrepancy. We conjecture a third possibility for constructing a quarterly measure of exports, based on firm data, and find that this does not work out. In the future, it may work, thus giving a 3rd measure of quarterly exports.

There are two independent sources of exports data:

1. RBI produces the Balance of Payments data. This obtains information about the exports of goods and services by watching flows of money through banks. This is released quarterly.
   
2. The Department of Commerce (DGCI&S) releases monthly data for exports of goods (only). This is based on the daily trade data that it receives from Customs authorities, SEZs and Ports. The phrase 'exports' is used when describing this data, but it is important to always refer to it as 'merchandise exports' or 'exports of goods'.
   

In order to assess the sanity of these two series, we compare the year-on-year change for them against each other.

The above graph compares the year-on-year change for the BoP exports and the merchandise exports series. Although varying in magnitude, the figure shows broad similarity in the ups and downs in the two series. The rank correlation between the two is 0.93. This seems sound.

The above graph plots the seasonally adjusted point-on-point growth (annualised) for the two exports series. This uses our work on seasonal adjustment. We observe similar trends in the two series. The rank correlation here is 0.83, which is mildly worrisome.

Can firm data yield a third measure of exports?

Large firms are important in India's exports, across both goods and services. As an example, in 2014-15, of the total exports of goods and services of USD 474.24 billion, the annual report data for 4666 exporting companies in the CMIE database reveals exports of USD 248.60 billion. Hence, we wonder: Could we use firm data to construct on exports time series?

The legal framework on reporting of annual financial statements requires that companies report their export income. Section 134 (3)(m) of Companies Act, 2013 states:

There shall be attached to statements laid before a company in general meeting, a report by its Board of Directors, which shall include: ---(m) the conservation of energy. technology absorption, foreign exchange earnings and outgo , in such manner as may be prescribed.

Rule 8(3)(C) of the Companies (Accounts) Rules, 2014 states:

The report of the Board shall contain the following information and details, namely-- ---(C) Foreign exchange earnings and outgo The Foreign Exchange earned in terms of actual inflows during the year and the Foreign Exchange outgo during the year in terms of actual outflows.

This clarity in legal drafting shapes the disclosure of export income of firms in their annual reports. A large chunk of firms disclose their export income in their annual reports. As an example as on 31st March, 2015, out of 4038 listed firms, 1795 firms are identified as exporters using this annual filing of financial statements. We could use quarterly data for these firms, which is required to be disclosed by all listed companies, to construct an exports measure. In terms of methods, we could do for exports what we have done previously for net sales using firm quarterly data.

When we turn to quarterly data, however, the legal instruments that define disclosure requirements do not clearly require information about exports. Under Regulation 33 of the SEBI (Listing Obligations and Disclosure Requirements) Regulations, 2015, SEBI requires companies to diclose their quarterly results, following the accounting standards prescribed by the Central Government under Section 133 of the Companies Act, 2013. The field on "Income from Operations" includes: (a) Net sales/income from Operations, and (b) Other operating income. (See Annexure I of this circular.) As a result, most firms do not report their 'export income' in their quarterly disclosures. Some firms voluntarily disclose information on exports, but this is part of the "Notes" to financial statements. As an example see Point 4 under Notes of this report.

In the table below, we present a comparison of firms that report exports in their annual reports versus those that report export income as part of their quarterly disclosures. The table shows the proportion of listed exporting firms that report 'export income' in their quarterly disclosures. It shows that a miniscule proportion of firms that are `exporters' as disclosed in their annual reports publish export income as part of their quarterly disclosures.

Year	Total Listed Firms	Exporters (As per AR)	Exporters (As per QR)	Share of firms
2000-03-31	3976	1891	54	2.9
2001-03-31	4373	2046	76	3.7
2002-03-31	4800	2013	97	4.8
2003-03-31	4876	1986	106	5.3
2004-03-31	4820	1986	108	5.4
2005-03-31	4990	1982	118	6.0
2006-03-31	4998	2037	131	6.4
2007-03-31	4884	2067	135	6.5
2008-03-31	4882	2084	101	4.8
2009-03-31	4893	2086	109	5.2
2010-03-31	4897	2046	88	4.3
2011-03-31	4867	2032	78	3.8
2012-03-31	4755	1996	49	2.5
2013-03-31	4612	1958	42	2.1
2014-03-31	4436	1911	43	2.3
2015-03-31	4038	1795	34	1.9

Recent improvements in the quarterly disclosures of firms

On 5 July 2016, SEBI has modified the format for publishing financial results under the SEBI (Listing Obligations and Disclosure Requirements) Regulations. This new circular mandates that the quarterly financial statements should follow the format prescribed in Schedule III to the Companies Act, 2013. Schedule III to the Companies Act, 2013 lays down general instructions for preparation of balance-sheet and statement of profit and loss of a company. Through requirements of 'additional information', companies are required to disclose income from exports. The relevant text from the legal instrument is as follows:

5. Additional Information The profit and loss account shall also contain by way of a note the following information, namely-- (e)Earnings in foreign exchange classified under the following heads, namely-- I. Export of goods calculated on F.O.B. basis;...

The revised reporting format becomes applicable for the period ending on or after March 31, 2017. From March, 2017 onwards we hope to see improved reporting by firms of quarterly exports data, which would make possible the construction of a third quarterly exports time series.

Conclusion

The Indian statistical system is riddled with problems of measurement. Critical pillars of the statistical system -- NAS, NSSO, ASI, IIP -- are not trusted. Every user of data must bring critical thinking into the choice of data that will be used. Our analysis above is reassuring in finding agreement between exports data released by RBI and the Ministry of Commerce, and rejects the possibility of constructing an exports measure using the present framework of quarterly disclosures by listed companies. Going forward, with improved reporting by firms we could get a third measure of quarterly exports.

 

The authors are researchers at the National Institute for Public Finance and Policy.

How to identify manufacturing companies for GDP estimation

by Amey Sapre and Pramod Sinha.

In a recent paper, we worked on the problems of gross value added (GVA) estimation for the manufacturing sector. One of the elements in the procedure is the identification of manufacturing companies from the MCA21 data. The GVA formula changes depending on whether a firm is classified as manufacturing or trading, and hence the classification of a firm into manufacturing or services is a critical question. We highlighted four concerns about the present procedures:

1. The use of reported ITC-HS codes can be misleading as the codes identify a product, and not a business activity.
2. For the purpose of GDP estimation, identification of companies has to be done every year.  In cases where the ITC-HS codes are unavailable, using the NIC digits in the Company Identification Number (CIN) can also be misleading. The CIN code does not change in time, and does not not track the evolution of the firm over time.
3. As the top revenue generating products of a company can vary yearly, this will require the statistical authority to identify and re-classify companies on a yearly basis.
4. In the absence of a feasible solution, wrongly classified companies will show an incorrect GVA contribution. On the aggregate, both manufacturing and services sector will show a distorted picture. These difficulties are compounded by the fact that the appropriate deflator to be used when converting nominal to real differs between the two cases, and has taken substantially different values in recent years.

Presently, the extent of distortion in the GVA estimate is unknown. In the paper, we try to estimate the extent of misclassification by looking for the two cases (i) firms that operate as non-manufacturing entities, but have their NIC codes registered in a manufacturing activity and (ii) firms that are into manufacturing, but have their NIC code registered in any other economic activity. However, we need to go beyond measuring misclassification to algorithms for better classification. In this article, we propose one such solution.

A potential solution

Currently, section II in the Form No. MGT 7, [pursuant to section 92(3) of the Companies Act, 2013 and Rule 12(1) of the Companies (Management and Administration) Rules, 2014] requires companies to furnish up to ten principal business activities . The information deals with disclosures of the main activity group, business activity with respective codes and their share in total turnover. Under this arrangement, the main activity has 21 different codes from A to U, each representing a particular activity. For example, a company reporting code C indicates a manufacturing concern, while code G shows trading. However, when non-reporting takes place, these codes alone will not solve the problem. A scrutiny of product schedules and financial statements is still needed.

Looking at product schedules and financial statements, manually, is expensive. This is particularly when low error rates are demanded. It is desirable to automate this work. We can see the rough contours of algorithmic classification as follows.

For a trading firm: Typically, for a trading company, from the revenue side, the income from trading to total turnover ratio would be higher than income from manufacturing. From the expenditure side, the ratio of purchase of finished goods to total expenses would be higher than the expenses on manufacturing.

For a manufacturing firm: In this case, from the revenue side, the ratio of income from sales to total turnover would be much higher than the ratio of trading income to total turnover. Similarly, from the expenditure side, the ratio of purchase of raw materials to total expenses would be much higher than expenses on trading. Also, for a manufacturing company, excise duty would be form a significant part of the indirect tax payments.

A statistical analysis of the ratios can help identify the characteristics of the manufacturing sector, and be used to classify firms effectively. The ratios can be applied to ascertain the highest revenue contribution on a yearly basis and at the same time allows a cross-check with reported codes and declaration under of Form No. MGT 7. The classification algorithm would need to deal with various categories of observation, including procedures that deal with various possibilities of non-reporting.



Amey Sapre is at the Indian Institute of Technology Kanpur and Pramod Sinha is a researcher at NIPFP.

The measurement of Indian manufacturing GDP: problems and some solutions

by Amey Sapre and Pramod Sinha.

Since the release of the 2011-12 series, the reliability of Indian GDP data has been the subject of intense debate. In the case of manufacturing GDP, there were large upward revisions in growth rates from 1.1% to 6.2% in 2012-13 and –0.7% to 5.29% in 2013-14, which were inconsistent with trusted private databases about growth in manufacturing. The introduction of the new MCA-21 dataset has also raised questions, as the lack of release has made it impossible for independent researchers to cross-check the estimates.

GDP estimation is a remarkably complex process. It is built on several sub-processes, datasets, and methodologies at the sub-sector level. At every base year revision, we see changes in sources and methods of computation that aim to yield improved measurement of macro aggregates. Valuable insights that can be derived, through the study of measurement issues, for our interpretation of the resulting data and thus our reading of macroeconomic conditions.

In a recent paper, we address three questions about Indian manufacturing GDP estimation:

1. Are we correctly measuring output and intermediate consumption in the formula for Gross Value Added (GVA)?
2. How sound is the technique of imputing missing data based on blowing-up using Paid Up Capital (PUC)?
3. When the new MCA-21 dataset is used, are manufacturing firms being correctly identified?

Questions about measuring output and intermediate consumption in the formula for Gross Value Added (GVA)

There are many concerns about how GVA has been estimated using firm data. In the paper, we recreate the process of GVA estimation. We use the Goldar Committee report in letter and spirit, and use the production side approach to recreate the GVA for a set of firms that file in MCA-21. For this, we take the XBRL formatted data from MCA-21 and identify the data fields used to compute GVA.

We also do a mapping of the XBRL fields with fields in CMIE Prowess and estimate the GVA. A detailed mapping can be found here. These two strategies give us a unique vantage point from which to evaluate discrepancies in GVA estimation.

Conceptually, the use of the MCA-21 dataset involves a shift from the erstwhile Establishment to the new Enterprise approach of value addition. The establishment approach captured production based data from factories registered under the Factories Act. The enterprise approach captures financial data of firms, and goes beyond just manufacturing by capturing value addition from post-manufacturing, ancillary or related activities such as marketing, and operations of branch/head offices. How does this impact upon value addition? There are two parts to this answer.

The first is the extent to which measures of output change.

Under the establishment approach, “Sales” was a measure of output. In the current enterprise approach formula, several disaggregated components of revenue that include revenues from products, services, operating revenues, revenue from financial services, rental income, incomes from brokerage and commission and other non-operating incomes are part of output. In the Goldar Committee report, there is a limited discussion on the inclusion or exclusion of several revenue fields in GVA computation. Also, the data labels and tags of the XBRL fields are broadly based on items in Schedule-III of the Companies Act. The lack of proper definitions of the fields makes the identification process cumbersome and prone to errors. It is evident from the output composition that value addition is not solely accruing from manufacturing activities, but also from several related activities. This leads to inflated GVA levels as the component of output is now similar to the total income of the company, and not industrial sales.

In the paper, we show a comparison with the previous sales based method and argue that changes in output composition alone can lead to increased levels of GVA. This will eventually push the growth rates upwards.

Comparison of GVA based on old and new method (Figures in Rs. Crore)

Year	Based on Sales	Based on Disagg -regated revenue	Difference
2011-12	701896.6	767311.4	65414.8
2012-13	742237.2	819228.5	76991.3
2013-14	780371.1	872178.1	91807.0

We study the firms in the CMIE Prowess. Using the traditional sales-based measure, our manufacturing GVA estimate is Rs.780,371.1 crore for 2013-14. Using the disaggregated revenue, it appears that there is an over-estimation of manufacturing GVA of Rs.91,807 crore, by including revenue from non-manufacturing activities.

What is missing in the GVA formula is a clear rationale of including revenues from different non-manufacturing activities. If such activities form a part of the enterprise level activities, it also requires a clear segregation of costs, identifiable data fields, and a consistent treatment in the formula to identify value addition from core manufacturing and other activities.

The second issue is changes in measures of intermediate consumption.

Identifying components of intermediate consumption at the enterprise level is equally difficult. Conventionally, subtracting the cost items (related to production) from output provides a measure of value addition entirely from manufacturing activities. However, with large and diversified enterprises, identifying cost items from financial data fields can pose significant challenges. A close scrutiny of the XBRL fields shows omission of important cost components, such as; Power & Fuel expenses, Advertisement and marketing related expenses. These are sizeable components and their omission can underestimate costs, thereby overestimating GVA. Thus, two possible reasons that account for distortion in GVA are; increase in output due to addition of several revenue items, and omissions in the components of costs.

Questions on the blow-up methodology

Missing data imputation is done, in Indian GDP estimation, by assuming that GVA is proportional to Paid Up Capital (PUC). In the paper, we replicate the blow-up process by constructing an available and active set of companies based on random samples that give different Paid-up Capital coverage. The details of the procedure have not been clearly documented in official publications. Several variants of the method are possible, such as; blow-up for each range of Paid-up Capital, blow-up by industry group, by ownership type of company, among others.

PUC-based blowup assumes that PUC and GVA have a deterministic and linear relation. This is at best a weak assumption, as one cannot draw sufficient inference about a company’s manufacturing activities by looking at its Paid-up Capital value. In the paper, we show that the size distribution of PUC and GVA have no systematic relation, and thus PUC is not an appropriate method to scale up GVA. Since the GVA contribution of a firm can be negative, the PUC based blow-up shows a distorted picture as it always contributes positively.

Our analysis of the blow-up procedure reveals several shortcomings. First, the blow-up factor is sensitive to Paid-up Capital (PUC) coverage and can show a considerable increase as the number of non-reporting companies increases. Second, the variation in blown-up values is unpredictable as there is no systematic trend for different values of the PUC factor. This leads to an unknown degree of error as the addition due to blow-up can be significantly large as compared to the actual contribution of unavailable companies.

On this problem, we are also able to offer a solution. In the paper, we show that using industry level growth rates of GVA to scale up previous year’s GVA of unavailable companies is a feasible and superior method. We use a sample to first classify each missing company into its industry and based on growth rates of GVA for each industry, we scale up the last available GVA of the unavailable company. Using industry level growth rates of GVA has an advantage over the PUC based blow-up as it uses the previous year’s GVA of the company instead of scaling up GVA of available companies. Industry growth rates capture the economic conditions faced by firms in and also provide a sufficient clue about the state of business environment. Computationally, on average, the method gives a lower margin of error, lesser variability, better representation of firm’s conditions and provides a close approximation to the actual GVA contribution of the firm. CSO can potentially shift to using this method.

Are manufacturing companies being correctly identified?

The Goldar Committee report makes a mention of using the ITC-HS product codes for identification of manufacturing companies. In absence of such codes, the Company Identification Number (CIN), which contains the NIC code, can be used to identify the nature of business activity of the company. The problems with using these two options are known. What is unknown is the extent of misclassification of companies and the error in the GVA estimate.

The reliance on ITC-HS code has several problems. Only 59% of the 30,006 companies filing in XBRL across all industries had reported the ITC-HS for products and NPCSS for services. However, even having the codes does not solve the problem. The codes only identify a product and do not distinguish between its trading and manufacturing. Thus, using such codes does not provide an assurance that value addition is being correctly captured for manufacturing products.

The problem is compounded in cases where the codes are unavailable. At present, a company’s CIN and the details on its website are used to identify its business activity. This yields misclassification.  For a company, its 21 digit CIN does not change once it has been created at the time of registration. Over time, a company may change the nature of its business activity or diversify into any other sector. This change of business activity is not reflected in the CIN code of the company. Using CIN can be potentially misleading since the top revenue generating activity of the company might be different from the one mentioned in its CIN code.

The NIC classification also changes from time to time. This adds to the complexity of identification in two ways; first, changes in business activities of companies are independent of changes in NIC codes, and second, a particular NIC code may not reflect the same business activity over time.

In the paper, we analyse this problem by studying two groups, (i) companies that operate as non-manufacturing entities, but have their NIC codes registered in a manufacturing activity and (ii) companies that are into manufacturing, but have their NIC code registered in any other economic activity. We show that there are a large number of companies in both categories and can create a significant distortion in the GVA estimate. We argue that any classification method for companies based on either ITC-HS or CIN code or hand mapping based on clues gathered from the name of the companies  or details from their website is likely to be incorrect. This process requires careful hand-analysis of each firm, to classify the firm correctly.

Conclusion

Sound computation of GDP is essential to decision making by the government and by the private sector. With imperfect observation of GDP, on many questions, we are flying blind. There has been a great amount of criticism of the Indian GDP data in recent years, as the high growth rates seen in the official data are inconsistent with trusted private databases. Our paper contributes three new blocks of knowledge to one component of the problem, i.e. measurement of manufacturing GDP.


Amey Sapre is an Economics Ph.D. student at IIT, Kanpur and Pramod Sinha is a researcher at NIPFP.

How bad is IIP growth after controlling for Diwali effects?

by Radhika Pandey and Pramod Sinha

Yesterday's data release showed a sharp contraction in industrial production in November. Some say that this contraction is an aberration as it was largely driven by seasonal fluctuations including the placement of Diwali and the number of working days in November.

A previous post on this blog How bad was industrial production in October? has highlighted the merits of using month-on-month seasonally adjusted numbers to provide timely and accurate information about the state of the economy. We apply our work on seasonal adjustment to analyse the true extent of contraction in the November IIP numbers. In the jargon of seasonal adjustment, Diwali is a `moving holiday': one that shows up in different months in different years.  A `Diwali effect' is found to be statistically significant for IIP and IIP (Manufacturing).

The figure above superposes the two time-series of seasonally adjusted annualised rates (SAARs). It has the SAAR IIP (without adjusting for Diwali) and SAAR IIP (with adjustment for Diwali). In most months, the two series are identical. But in some months, the interpretation of the IIP data strongly requires adjustment of Diwali as a moving holiday. The numbers for the month of November 2015 are worth noting. Without adjusting for Diwali effect, the month-on-month growth shows a sharp contraction of -71%, after adjusting for Diwali effect the month-on-month numbers modestly improve to -39%.

The figure above shows the same analysis for IIP (Manufacturing). If we look at the manufacturing sector, the seasonally adjusted month-on-month growth yields a negative growth of -44%. Adjusting for diwali effect results in modest improvement to -33%.

To conclude: After adjusting for the Diwali effect, IIP dropped by an annualised 39%, and IIP manufacturing dropped by an annualised 33%.


The authors are researchers at the National Institute for Public Finance and Policy.

A better output proxy for the Indian economy

by Akhil Dua, Pinaki Mukherjee, Radhika Pandey, Ila Patnaik, Pramod Sinha, Ajay Shah.

India's emergence as a market economy has been accompanied by the emergence of business cycle fluctuations that are similar to those seen in market economies [link, link]. In understanding business cycle conditions, and in crafting institutional arrangements for stabilisation, it is essential to properly measure output and prices.

The problem

India is in reasonably good shape on measurement of prices, with older CPI-IW data and now the new CPI. (Using WPI as a measure of prices is wrong, but you don't have to make this mistake; the statistical system does have CPI-IW and then CPI). Measurement of output, in contrast, has presented serious difficulties.

The index of industrial production is widely used as a measure of business cycle conditions. However, it reflects only manufacturing, which is a small part of the economy. In addition, it is riddled with serious difficulties of measurement.

The other key measure that has been widely used is quarterly GDP data. However, contrary to what textbooks teach us, in India, quarterly GDP data is constructed without information about the demand side. In addition, there are two important concerns about the quarterly GDP data from the viewpoint of business cycle analysis:

• Agriculture is included in the overall GDP data -- as it should -- but to a significant extent, fluctuations in agriculture reflect weather shocks and do not reflect underlying business cycle conditions.
• Spending by the government is counted as output in GDP data. However, it does not reflect underlying business cycle conditions. See Robert Higgs on this subject.

As a consequence, quarterly GDP data in India is not a good reflection of business cycle conditions.

Two steps towards measuring output

In order to address these issues, we have constructed two new series which, we feel, do a good job of measuring nominal output.

The first of these is GDP excluding agriculture and excluding government. This focuses upon the output of individuals, small firms and large firms, which is what the market economy and the business cycle is all about.

The second strategy consists of utilising firm data. Listed companies are required to release quarterly results. These results are pored over by accountants, auditors, senior managers, tax collectors, shareholders, etc. They are thus likely to have few mistakes of the sort which have plagued government statistics and survey-based information.

Finance companies have very different concepts underlying their accounting data. In particular, banks are large firms which overstate their operating performance when bad loans are hidden. Hence, financial firms are excluded. Oil companies sometimes experience very large jumps in their revenues owing to decisions by the government about administered prices. These fluctuations are not a feature of underlying business cycle conditions. Hence, oil companies are excluded. In short, we focus on all listed firms observed in the CMIE database other than finance and oil companies.

For each pair of quarters, we construct a panel of firms observed in both quarters, and work out the percentage change in the sum of net sales across all the firms. These percentage changes are used to construct a net sales index.

Non-agricultural and non-government GDP is the business cycle. It is made up of production by small firms (going down to one employee) and large firms doing both industry and services. This measure captures large firms in both industry and services and is thus a good proxy for what is going on in business cycle conditions.

Net sales of non-finance non-oil firms, and GDP ex-agriculture and ex-government.
Nominal indexes, non-seasonally adjusted.

The graph above shows these two time-series. The fact that the two series -- which are constructed from completely unrelated sources -- agree with each other across long periods of time is a source of increased confidence.

Net sales of non-finance non-oil firms, and GDP ex-agriculture and ex-government.
Nominal indexes, seasonally adjusted.

 Our first step is to seasonally adjust both series. Once again, it is satisfying to see how well the two series agree with each other, even though they are quite unrelated on their underlying sources.

2-Q moving average of growth of seasonally adjusted levels.
Annualised per cent.

Using this, we are able to compute nominal GDP growth. Once again, it is striking how well the two series agree with each other. Two major recessions are visible: 2001 and 2009.

This series shows much more macroeconomic volatility when compared with what we are used to with conventional data. This is perhaps unsurprising as government expenditure is a fairly stable series, and fluctuations of agriculture are noise. When these two are removed we see substantial macroeconomic volatility. This is not surprising, as India presently lacks the frameworks of stabilisation through either monetary or fiscal policy.

Conclusion

We feel that we now observe a good measure of output in India, with two different measures that are gratifyingly close to each other. These series are a valuable starting point for business cycle research.

The key flaw of this work is that it takes us till a nominal output index. The next big hurdle to cross is that of converting to real. The simplest strategy would be to just use the CPI-IW and then the new CPI.