DIY assignment: Off-balance sheet assets in China’s banking system

China’s Financial Stability Report for 2016 published in 2017 is now available in English. You can download it from here. It does not show up under the tab, ‘Financial Stability’. Under that tab, you can download the 2016 report which covers the year 2015.

In the 2017 report, go to the Chapter on Banking Sector (page 48 as per their page number; not as per the PDF file. As per the PDF, the page number is 64) and understand the figures of assets on and off-balance sheet in the China Banking Sector.

Do the same thing with the 2016 report. Go to page 53 of the document. That is page 69 of the PDF document. Find out the off-balance sheet asset size.

The on-balance sheet asset size comes at the beginning of the Chapter in each of the reports.

You can also get the nominal GDP of China from one of the earlier sections.

Now you are ready to do your math on China’s banking system assets and their ratio to GDP, etc. (on and off-balance sheet).

You will know why we have a ‘recovery’ in 2016.

This is what I was alluding to (without proof but with some intuition) in my MINT column which wondered whether the global recovery was statistical or real. Well, I should have asked if it was financial or real.

A report by Macquarie Research (not available in the public domain) released few days ago is cynical but also correct. It backs the perennial doomsday machine to continue because, without that, financial assets worth some USD400trn will be exposed for their hollowness as they are backed by real assets worth far less or nothing.

It is liquidity and leverage or nothing.


Andy Mukherjee’s India chessboard

The advantage of reading Andy Mukherjee’s Bloomberg Gadfly columns in a bunch is that one is able to form a string of pearls with them. The chronological distance gives some perspective.

Some of his recent pieces are:

Ratan Tata turns his holding co. private [Link]
Incredible tax demand on Docomo settlement money [Link]
Private sector Boards not firing their non-performing CEOs [Link]
Only achievement of GST being to deprive exporters of their working capital [Link]
Synergies-Dooray made a meal of the IBC [Link]

The chess board does not look very pretty. The King (Indian Economy) is increasingly besieged.

Working Capital for small businesses in India

For Small and Medium Enterprises (SME), a big issue is the availability of working capital and the cost of financing it. See the earlier blog post on the productivity of working capital.

SMEs that supply to big corporations are usually at their mercy for payment of invoices. If the buyer is monopsony, then the supplier’s bargaining position is even weaker. Nor can they afford to pursue legal remedies against recalcitrant corporate customers. Such remedies will be costly, protracted and may ruin the business too. ‘Factoring’ is the answer. SME suppliers discount their receivables and bills of exchange with banks who buy them (with or without recourse to the SME, in the event of non-payment) at a discount and then collect the proceeds from the corporate buyer.

The idea was included in the report of the Raghuram Rajan Committee on Financial Sector Reforms in 2008. Then, RBI appointed a working group which submitted its report in November 2009. RBI then issued a concept paper in 2014. Then, the draft guidelines were issued followed by final guidelines. Receivables Exchange of India (RXIL), India’s first Trade Receivables Discounting System (TReDS) started operating only in January 2017.

The time taken underscores the glacial pace at which crucial reforms that would lift the economy on to a higher growth path are pursued.

The world over, on providing working capital to small suppliers, things are moving much faster (ht: Gulzar).

Six months later, hurdles to the smooth of functioning of RXIL are emerging. Companies are wary of uploading invoices lest competitors come to know about their suppliers. Second, this is important, “since TReDS is a transparent system, they (companies) necessarily would have to settle the suppliers’ invoices within 45 days of acceptance of goods/services rendered.”

From a policymakers’ point of view, starting from the Panel Discussion on Agriculture and the posts on India’s informal, small and large factories, three interventions are becoming clearer:

(1) Risk mitigation in the form of a functioning crop insurance market with the Government bearing the premiums payable to cover not only yield, output but also realisation risks.

(2) Irrigation coverage – canals, inter-linking of rivers (long-term and huge capital commitment), rainwater harvesting (ongoing and needs top-level ownership like in the case of ‘Swachh Bharat’)

(3) For informal enterprises, financing via Mudra Loans but with performance caveats

(4) For small enterprises, working capital access – discounting, factoring, etc.

(5) For small and informal enterprises: revamp, re-design and re-energise the Skilling Initiative. Make contributions to Skilling CSR Compliant (if not already done). Tie it up with big corporations. I do not know if there is further work to be done on the Apprenticeship Act.  It has been passed. Have the rules been framed under the Act, been notified and circulated?

(5) For all: stable, reasonable tax regime that is non-vindictive and non-usurious in its administration. An immediate fallout of it is GST rationalisation of rate structures and further lowering of rates. Bet on economic activity and not on economic policing.

(6) Government to expend fiscal resources on bearing more labour payroll deductions for entrepreneurs (increase the salary cut-off limit for government reimbursement)

(7) Follow through on corporate tax rates reduction with elimination of ad-hoc exemptions

Without any trace of immodesty, I would venture to state that if the Indian Government could focus on these seven for the next eighteen months, it will have done a lot to relieve the economy of the strains and stresses it is facing now.

Productivity of Indian factories – small and big

Defining productivity is one of the hardest things on economics. A well-known economist-journalist reminded me of it when I used that word in an email and he asked me for my definition. It is hard. It is a bit like beauty and obscenity. You know it when you see it. Of course, it is not as hard as those concepts. In factories, labour productivity can be measured. But, capital productivity is somewhat more problematic and low productivity may not be entirely for internal reasons.

Suppose the construction of a factory begins in year 1 and some capital is committed. But, let us say, approvals take forever to come – environmental, legal, utility connections, etc.  Here, capital has been committed and invested but yet, there is no production to show. There is unproductive capital here. But, one cannot say that it is the fault of the enterprise or the entrepreneur.

Total Factor Productivity (TFP) is even more complex.  It is the residual. Output growth that cannot be explained by input growth is TFP.

Further, in the case of India, with services dominating the economy, it is even harder to get a grip on productivity. Measuring productivity in service sector could be problematic as there could be a conflict between the micro and macro on this. This is not the place to elaborate it but more on that on another occasion.

It took a laborious (for me) blog post to establish that units/factories covered by the Annual Survey of Industries are not ‘informal’. They are registered under the Factories Act, 1948 under Sections 2 (m) (i) and 2 (m) (ii). They are not covered in other surveys of unincorporated and unorganised enterprises by the NSSO, etc.

Therefore, productivity of the smaller factories – to be presented below – is not the same as productivity of informal manufacturing enterprises.

I prepared two Tables on productivity statistics for Annual Survey of Industries (ASI) – Factories. The latest ASI is for 2014-15 released in March 2017. The tables below are based on statistics presented in Statement 11-A (‘Principal Characteristics by the Size of Employment’ – Section 7, page no. S7-3).

Productivity statistics for Factories with 99 or less workers

Productivity statistics for Factories with 100 or more workers

The Tables are along expected lines.

(1) Capital productivity of smaller units is better because they deploy lesser capital. In a relatively capital scarce country, that is better. But, is it true that India is still as capital starved as it used to be, with capital flows being global and Indian savings rate being upwards of 30%? In fact, how much of the capital scarcity is because Governments capture a large portion of the national savings through the banking system?

(2) Labour productivity of larger units is much better because they employ fewer workers relative to the output or GVA they generate.

(3) What is interesting is that on capital productivity measures, smaller units are not decisively better off.  GVA/Invested Capital Ratio is the same for both small and big factories. Productivity of working capital is far better for larger factories. Of course, that tells us how big a constraint working capital is, for smaller enterprises. That will be my third post.

(4) Output/Wage Ratio (in Rupees) is better for smaller units. May be, because they do not pay well? But, GVA/Wage ratio is much better for larger factories.

(5) Similar is the story with respect to Emoluments.

(6) Overall output/input ratio is better for larger enterprises.

These two tables should help us design appropriate policy intervention. Some conclusions that emerge, in my view:

(1) India needs both large and small factories. In fact, small factories arise for the most part as part of the supply chain for larger factories.

(2) Working Capital Efficiency for smaller factories can and should be improved. The onus is on the Reserve Bank of India and on larger corporate customers

(3) Smaller units need labour skills and training

(4) Larger units need to improve their capital and labour productivity but not to beat their smaller counterparts in India but to compete globally. They must benchmark themselves globally.

(5) Clearly, the data do not make the case for the country to ignore the need to allow enterprises to grow big. There is no case for a ‘anti-big’ bias. In fact, if anything, ‘informal’/’unorganised’/’unincorporated’ enterprises may have even poorer productivity statistics than the small factories covered by the ASI.

Organising my thoughts on the unorganised sector

The Sixth Economic Survey was conducted in 2013-14 and the report was released in March 2016.  You can find it here.

On page 13, it says:

Economic Census (EC) is the complete count of all establishments (i.e. units engaged in the production and/or distribution of goods and services not for the purpose of sole consumption) located within the geographical boundaries of the country. The Sixth EC was conducted during January, 2013 to April, 2014 (see Annexure V for details) in all the States and Union Territories of the country in collaboration with State/UT Governments.

All economic activities (agricultural and non-agricultural), except those involved in crop production and plantation, public administration, defense and compulsory social security, related to production and/or distribution of goods and/or services other than for the sole purpose of own consumption were covered.

On page 25, it says:

As per the results of the Sixth Economic Census (Table 2.1), there are 58.50 million
establishments in the country engaged in different economic activities other than crop production, plantation, public administration, defence and compulsory social security services.

Out of which, 34.80 million establishments (59.48%) are in the rural areas and 23.70 million establishments (40.52%) in the urban areas.

But, here is where the confusion arises. Page 13 also states the following:

Based on the frame thrown up by Fifth Economic Census, the following follow-up surveys were carried out:-

(i) National Sample Survey (NSS) 67th round during 2010-11 (Survey on Unincorporated Non-Agricultural Enterprises excluding Construction);

(ii) NSS 73rd round during 2015-16 with the coverage similar to that of NSS 67th round.

The full report of the National Sample Survey Organisation (NSSO) 73rd round is available here.  A summary is available here. It shows that India has an estimated 6.34 crore non-agricultural, unincorporated enterprises (same as  ‘Establishments’?).

It should be a subset of the overall ‘Establishments Survey’ of the Sixth Economic Census because that survey included agricultural establishments too.

In fact, the introduction to the Report 73rd round of the NSSO brings out the difference between its findings and that of the Sixth Economic Census: While the 6th Economic Census covered all the unincorporated enterprises as included in the coverage of NSS 73rd round, it also covered all other units engaged in various agricultural and non-agricultural activities excluding crop production, plantation, public administration, defence and compulsory social security. While covering these activities, 6th Economic Census considered the following ownership categories for inclusion:

 Government/ PSU
 Proprietary and Partnership establishments
 Private Corporate Establishments (Companies)
 Non-Profit Institutions (NPIs)
 Trusts
 Cooperatives
 Self Help Groups (SHGs) The SHGs, which were formed for engaging in financial intermediary services and later changed into some group based non-financial activity, were also considered as SHGs in 6th EC. However, units formed as an SHG and engaged in non-financial activities were considered as a partnership enterprise with members not all from the same household. All members of SHG who were regularly attending meetings or taking part in decision making procedure like secretary, treasurer, active committee member etc. were treated as working owners in 6th EC. However, they were not considered as working owners (or workers) in NSS 73rd round.

So, the Sixth Economic Census must have included a much larger set than that of the NSSO 73rd Round. But, the number of Establishments, according to the Sixth EC is only 58 million. The 73rd Round already mentions 60.34 million unincorporated enterprises. The Sixth Economic Census is already outdated?

Why use two different words, ‘Enterprise’ and ‘Establishment’? The definition of ‘Establishment’ given above does not seem to indicate that it only covers ‘unincorporated’ Establishments.

India uses the words, ‘small’,  ‘unincorporated’, ‘informal’, ‘unorganised’ rather interchangeably. For the most part, they overlap but there could be critical differences. Policy interventions would be more effective if the understanding is clear and common across all stakeholders and decision-makers.

According to this Wikipedia entry, India defines the ‘unorganised’ sector as comprising of

all unincorporated private enterprises owned by individuals or households engaged in the sale or production of goods and services operated on a proprietary or partnership basis and with less than ten total workers.

So, ‘unorganised’ is ‘unincorporated’. Now, let us turn to ‘informal’. C.P. Chandrasekhar, in this article in THE HINDU in September 2014 cites the definition of the NSSO on ‘Informal Sector’:

Proprietary and partnership enterprises (excluding those run by non-corporate entities such as cooperatives, trusts and non-profit institutions), in the non-agricultural sector and in agriculture-related activities excluding crop production (AGEGC).

So, ‘informal’ is also ‘unincorporated’ is also ‘unorganised’.

But, ‘unincorporated’ does not mean ‘unregistered’ because the NSSO 73rd Round Report included ‘Manufacturing enterprises registered under Section 85 of Factories Act, 1948’. It has not included those factories that are covered by the Annual Survey of Industries (ASI) since that survey only includes those Factories that are covered by the definitions as per Section 2(m)(i) and 2(m)(ii) of the Factories Act, 1948.

The NSSO 73rd round has explicitly excluded those, as you can see below:

The survey covered the following broad categories:
(a) Manufacturing enterprises excluding those registered under Sections 2m(i) and 2m(ii) of the Factories Act, 1948
(b) Manufacturing enterprises registered under Section 85 of Factories Act, 1948
(c) Enterprises engaged in cotton ginning, cleaning and baling (code 01632 of NIC-
2008) excluding those registered under Factories Act, 1948
(d) Enterprises manufacturing beedi and cigar excluding those registered under beedi and cigar workers (conditions of employment) Act, 1966
(e) Non captive electric power generation, transmission and distribution by units not
registered with the Central Electricity Authority (CEA)
(f) Trading enterprises
(g) Other Services sector enterprises excluding construction

Categories of enterprises under coverage in (a) to (g) above were:

(a) Proprietary and partnership enterprises [excluding Limited Liability Partnership
(LLP) enterprises]
(b) Trusts, Self-Help Groups (SHGs), Non-Profit Institutions (NPIs), etc.
Following enterprises were excluded from the coverage:
(a) Enterprises which are incorporated i.e. registered under Companies Act, 1956
(b) The electricity units registered with the Central Electricity Authority (CEA)
(c) Government and public sector enterprises
(d) Cooperatives

In other words, these  NSSO Surveys cover establishments that are truly unorganised, unincorporated and informal. Whereas Factories covered the ASI are not part of these.

The Economic Census is a different ball game. It is a census of all establishments including incorporated enterprises and government/public sector undertakings.

Therefore, when one is evaluating the productivity of factories covered by the ASI, one is comparing productivity of firms within the so-called ‘formal’ sector except the sizes may vary.

I had to do this blog post just to organise my thoughts in my head clearly and to set the stage for comparing productivity of Factories as covered by the ASI. That is the next blog post.

Nayakan and Interest rates on small savings deposits

I just managed to go through the RBI Monetary Policy Report released early in October. The RBI does this twice a year. Some interesting discussions, as always. I shall refrain from copying and pasting text as much as possible but give my views with reference to the page numbers. There are 78 pages in the PDF file.

RBI has several and more substantive upside risks to inflation than downside risk to inflation. (Page 14)

Discussion on the collapse in the prices of pulses is useful. RBI points to information dissemination failure. But, it is also mentioned twice that the procurement operations fell short. Quite why State governments and FCI failed to procure even though the government had increased the Minimum Support Prices for pulses is a question that remains unanswered.

Is it execution problems or is it because of concerns over fiscal deficit that made State governments go slow on procurement? But, now they are not saving anything on the fiscal side with their farm loan waivers. (Pages 24-26)

RBI has sharply rising Gross Value Added (GVA) growth rates in the second, third and fourth quarters of the current fiscal year, 2017-18: 6.4%, 7.1% and 7.7%. This is notwithstanding the Chart III.10 in page 42. GVA excluding Public Administration, Defence and Other Services is growing a lot more tepidly. For 2018-19, the estimate is 7.4%, assuming normal monsoon and fiscal consolidation (Page 17). Considering tepid investment demand and farm sector troubles (see point below on reservoir levels), the sharp acceleration in GVA growth penciled in appears quite optimistic. May they be right!

Chart II.20 is a bit confusing. It has two panels. Panel A shows that growth in ‘Per Employee Cost in the Organised Sector’ in Services is declining. But, Panel B shows that ‘Staff Cost per unit value of production’ in Services is 27% and has increased from around 15% in 2009. I do not know if it is growth rate in the Staff Cost or is it ‘Share of Staff Cost Per Unit Value of Production’. (Page 34).

Basing itself on CMIE data, the central bank says that new investment proposals are at their lowest since 2004-05 (Page 37). It is also borne out by the capital expenditure outlays that it puts out every year in its September monthly bulletin, based on filings made by borrowers and fund-raisers for capital expenditure.

Further, in its Annual Report, the central bank had said that stalled infrastructure projects (number and value) had shown a material decline. See my post here. In the Monetary Policy Report, it points to anecdotal evidence on road construction being at an all-time high. (Page 38).

Farm loan waiver in UP has extracted a huge cost in terms of decline in capital outlay and capital expenditure. (Page 40).

Water storage level in 91 reservoirs across the country were about 66% of the full reservoir levels (as opposed to 74% last year). In fact, apart from the Northern Region, reservoirs in all other regions received less water than in the corresponding period last year. (Page 44).

Growth in GVA in manufacturing enterprises is -5%! (Chart III.19 – page 45)

Pity that the central bank does not mention its estimate of potential GDP (or, GVA) growth in its box on Output Gap. Is it closer to 7.0%? (Page 49).

Discussion on monetary policy transmission is good. Only 75 basis points of reduction in the base rate against a policy rate decline of 200 bp. New loans benefited more than outstanding loans! Think of it, it is a bit unfair to your loyal customers who are paying regularly.

Interesting (and troubling too?) that housing loans and loans to large industry are priced competitively and in line with one-year median MCLR as well as the base rate (Pages 62-63).

Table IV. 4 in Page 64 evokes mixed reactions. Is it good or bad? Like the grandson who asks Kamal Hassan in ‘Nayakan’?

Under the complex economic reality that has evolved in the country or been allowed to evolve, by the government, with high fiscal spending (more on loan waivers and less on capital expenditure and irrigation, for example), high inflation, etc., this becomes inevitable? Is it a third or fourth best world?

A brief panel discussion on the state of Indian agriculture

The discussion arose out of the news-story in THE HINDU on Rahul Gandhi promising farm loan waiver in Gujarat.

Panellist 1: Actually agriculture is a pretty complex system and conventional market solutions do not appear to work. Gluts and shortages are inevitable – bad weather is a risk; good prices lead to excess cultivation next year and resultant drops, and vice-versa; global supply shocks and resultant price fluctuations are always round the corner; poor storage and other forward linkages make farm sales the only option etc. Pain and suffering follows.

Developed countries, over decades, have sought to address this problem through less distorting approaches – mainly crop insurance and/or direct payments. It helped that they have good irrigation systems, farms are bigger, forward linkages are better, credit access simple, and markets are functional.

We have none of the positive conditions, and crop insurance and direct payments are both very expensive and run into problems of effective administration.

But we have this smorgasbord of inefficient and distorting things – subsidised crop loans and their recurrent waivers; procurement and MSP (which feeds into the PDS); fertiliser subsidy; free farm power; agriculture IT exemption; minor irrigation programs like PMKSY etc. Worse still, each one has generated its set of powerful entrenched interests, which reflexively gang up as a vocal electoral constituency whenever they are threatened. Making matters complex, it cannot also be denied that each one of these, in their very sub-optimal ways, contributes to mitigating, even if partially, the fundamental problem, and the resultant pain and suffering.

So we have a very bad self-reinforcing and perpetuating equilibrium. A ‘chakravyuha’, from which exits appear very daunting…

I just have not come across anything satisfactory as a path out of this. Except the gradual process of development – build irrigation systems, transition people out of agriculture, consolidate farms, let linkages and markets develop etc. – and the gradual introduction of things like crop insurance. In this dismal environment, doubling farm incomes may well be the government’s most over-optimistic promise yet.

Panellist 2: Thoughtful response. I doubt if there is much to add to what you have written, since it encompasses all angles – what needs to be done, what is feasible, what needs to be worked on and what is the role that even the so-called sub-optimal solutions play in mitigating the seemingly insurmountable issues surrounding agriculture in India compounded by India-specific issues.

May be, one small thought. You say that ‘Crop insurance is expensive’. So, instead of these expensive and recurring farm loan waivers – won’t it be a more effective application of government subsidy if the government were to bear the premium. Of course, government paying premiums to, say, a private insurance entity attracts all kinds of criticisms. But, can one design it to be ‘as above board as is possible’ and pay the premium?

Panellist 1: Among all the solutions I think what you are suggesting may be one of the best options. Given that the crop insurance scheme is one of the government’s bigger initiatives, and one of the most progressive (as well as efficient), I think they should actually spend more energies and resources on it. Premium support will always entail big subsidies and this may be worth paying. Can the government think of phasing out some of these other subsidies and phasing in more of crop-insurance support? For example, it could encourage states which are willing to do farm power metering and a low agricultural tariff (to start with), to be provided a much higher premium support subsidy…

Panellist 2: I am glad that, preliminarily, you seem to find the idea the most practicable. As you had written, it can be tied to some other improvements (better not call them ‘reforms’).

Panellist 3: There is one optimistic scenario where farm incomes can double despite a continuing economic mess in the sector. That is if something like 1965-70 happens in terms of technology. Dryland agriculture –untouched by the earlier Green Revolution–now has some promising yield enhancing techniques which are not fully ‘extended’; similarly the SRI system offers large benefits in paddy cultivation. In both cases, the broken agricultural extension machinery is not able to propagate them adequately–whereas in the 60s/70s the extension system worked. Similarly, a quick expansion of the cold chain can quickly increase the share of income going to farmers and reduce their vulnerability by giving them staying power. These do not require politically difficult measures–just good administration but that is not easy either.

On crop insurance, there is an interesting idea from Dr. Ramesh Chand, member NITI–he suggests abolishing it (in its present form), instead announcing free government subsidies when crop losses occur, in an amount equal to what would be given under crop insurance. No time to discuss it here, but in terms of fiscal outgo, his calculations show it can actually reduce costs, by eliminating the transaction cost of dealing with ALL farmers and replacing them with dealing only with those who make losses. He makes the point that for the same overall fiscal cost, Govt. can give it ‘free’ and earn brownie points rather than appear to collect a premium that is a fraction of the actuarial cost. It would also help reduce the rampant insurance frauds that are witnessed in some States.

Panellist 1: About SRI, I am not sure. It has been around for long. The last I heard was that it had expanded but not by much. Also, it has been heavily researched and the evidence has been mixed

Panellist 2: One quick comment: “announcing free government subsidies when crop losses occur, in an amount equal to what would be given under crop insurance.”

If a government compensates only one set of farmers and not others, no matter how objectively done, will it be accepted without protest? All farmers, for example, now want loan waivers, regardless of whether they are distressed. The Madras High Court backed them, in Tamil Nadu!

If an insurance company pays out on the basis of verified loss claims, it is far less controversial. My two cents worth.

My comment:

If you noticed, the emphasis is on risk management and risk mitigation. How to design and a run a system of insurance, taking Indian context and potential difficulties – political and administrative – into account. But, crop insurance has to be a priority.