Crude oil consumption in China grew while the world total didn't

A few days ago, James Hamilton had a blog post on the global oil market. He puzzles over this set of facts: From 2005 to 2007, Chinese crude oil consumption went up by 860,000 barrels/day, while global output dropped by 30,000 barrels/day. How could this have happened?

One element of an explanation could be like this. We know that China produces a lot of widgets. Chinese market share in world production of widgets has been growing fast (e.g. see this graph with growth of Chinese exports to the US). The production of widgets requires energy. Chinese `market share' in energy consumption should rise with the gains in Chinese `market share' in manufacturing production. As manufacturing work shifts to China, it would look like Chinese GDP is incredibly energy-intensive, while other locations would seem less energy intensive.

By the above calculations, China's consumption went up from 8% of world crude oil output to 9% from 2005 to 2007. That's roughly consistent with Chinese share in world merchandise exports.

Responding to the credit policy announcement of yesterday

Inflation, growth

I have an article Parsing the credit policy in Financial Express today in response to RBI's credit policy announcement yesterday.

How can we change this boom and bust cycle?

Towards the end of this article, I argue that we need to see that this kind of shock therapy is not what a mature market economy does. In recent years, we have gone through a sequence of first having loose monetary policy (capital inflows + exchange rate pegging); this ignited a nasty inflation; now we are going through a belated response of raising rates in order to combat this. None of this would happen with a properly structured central bank. With monetary policy reform, we would substantially change the boom and bust cycle that afflicts India.

There is no short rate, there is no monetary policy rule

I am mystified by the 300 bps gap between the repo and reverse repo rates. Why would this be optimal? Suppose the government starts spending (or if RBI buys USD). The call rate will then suddenly crash from near 9% to near 6%. This is no good in terms of fighting inflation. In addition, it makes the job of the financial sector in general and banking in particular extremely hard.

The task of monetary policy is (a) to pin down the short term rate, and (b) to write down and transparently communicate a monetary policy rule, so that expectations about future values of the short rate can then trace out a yield curve. In India we are doing neither. There is no monetary policy rule. There is no short-term policy rate - there is a yawning gulf of two policy rates separated by a 300 bps zone of uncertainty.

Such behaviour on the part of the central bank comes at a cost to the economy. It is not surprising that we then have problems obtaining liquidity on the bond market. Ultimately financial firms including banks charge households and firms in the economy more, to pay for these risks and difficulties. Bad monetary policy induces bad financial intermediation.

RBI regulated entities have no freedom of speech

Watching bankers on television was distressing. CRR had just been raised - this is a tax on banking. Banking stocks had just been damaged: the CMIE Banking services stock price index dropped 7.92% on a day when the broad market fell 3.04%. But bankers on television were above all this. They were praising the wise RBI for having been drafted a Most Optimal Monetary Policy. It was a scene out of a communist country. As the Indian Express editorial says:

However, one unfortunate element of the regulatory regime in Indian banking, clear from the coverage of credit policy, is how little freedom of speech is afforded bankers. With the hike in the CRR, banks have to put additional resources in non-remunerative deposits with the RBI and this hits their profits. When in response to the policy statement bank stock prices were falling, most bankers were standing by saying that they agreed with everything that is being done. Few objected to the announcement that the RBI would set up audits and committees to control the credit growth of banks. In the licence and permit raj that rules the Indian banking system few are willing to say openly how this would affect them.

Auction-based procurement for fund managers is working well

A key development that has been taking place in India is the translation of the idea of auction-based procurement of fund managers into reality. Mainstream finance practitioners do not like to bid in auctions focusing on fees and expenses; it is argued that if you pay peanuts you will get monkeys. But the finance research literature has supported this kind of quest, emphasising low-cost passive management as the investment strategy of choice when dealing with liquid securities.

This idea was first proposed in India by the Project OASIS committee (who, in turn, got it from Estelle James). Now we have three success stories of such bidding:

Coal Mines Provident Fund, March 2007

1 basis point. This was 5x cheaper than the erstwhile arrangement.

New Pension System, 27 November 2007

3-5 basis points.

EPFO, 29 July 2008

0.63-1 basis points. Note that the erstwhile monopoly - SBI - was not the lowest bidder by far. EPFO has given assets to fund managers based on the strength of their bids.

Gautam Chikermane has a good piece in Hindustan Times sketching these events and their larger implications for pension reform. Also see an optimistic editorial in Indian Express.

There has been a certain amount of criticism of the procurement proces. See this gloomy edit in Financial Express, the story from Shaji Vikraman in Economic Times, and this story by Vikas Dhoot in Financial Express. I am obviously not aware of the full facts. It is true that the Ministry of Labour and EPFO are relatively inexperienced with such procurement. They face a hostile environment in the Central Board of Trustees. Despite these constraints, by and large, in my opinion, the outcome was a good one. I don't share the perspective of these critics.

When the Project OASIS report came out, many finance practitioners claimed that this report was out of touch in reality when talking about rock bottom prices for fund management using auctions. Now we have three auctions in hand, and it's safe to say that the fair price for wholesale fund management in India is atmost 5 basis points. While we don't yet have economies of scale in Indian fund management, we do have lower wages, which is giving us world-class prices even though the scale economies are not like those seen internationally. I believe that in a few years, the thumb-rule in India will be "the fair price for wholesale fund management is atmost 2 basis points" owing to bulking up with economies of scale. To help you fix ideas, an AUM fee of 2 basis points on an AUM of Rs.10,000 crore is Rs.2 crore a year which is about all it costs to run passive portfolios.

While on this subject, do see the `Wholesale Asset Management Companies' proposal of the Percy Mistry report (para 56-58, page 202 onwards).

If you are an individual investor using mutual funds or insurance companies, you should be weeping about the prices that you are paying. For a personal investing perspective, see this. From the viewpoint of the economy, this underlines the cost-effectiveness of a market-dominated financial system: it is able to link up wholesale liabilities into assets while suffering roughly zero friction.

Closure of financial firms

Business Standard has an excellent editorial on the problems of Sahara India Financial Corporation. It talks about the difficulties of shutting down financial firms.

Raghuram Rajan's report emphasises the importance of a mechanism for swiftly shutting down insolvent financial firms. For a role model, last weekend, the US FDIC closed down two banks. Notice how swiftly and smoothly these closures were done. Here is the stock information release that FDIC puts out about such an event and here's the press release that they put out.

Compare and constrast this with the ineffectual DICGC that we have in India. After 15 years of talking about financial sector reforms, we have not moved an inch on this.

Two consequences follow when closure of financial firms is made difficult. The competitive landscape is contaminated by the perpetuation of `zombie firms' who should be dead but aren't. These firms block capital and labour that can be more productively used by other firms in the economy. And, they reduce profit rates for healthy firms and adversely affect investment in the economy.

In India, difficulties of closure of financial firms is part of the package of excuses that is offered for blocking progress. E.g. it is claimed that derivatives trading cannot be allowed to take root because some firms will experience distress when making mistakes with derivatives, and since that distress is painful for the regulator, all firms should be denied the benefits of derivatives trading.

One of the crucial elements of the policy package which has worked out right for the equity market is swift closure of firms in distress. When NSCC sees a firm which has violated rules about margins, its trading permissions are immediately rescinded (even if this involves triggering off panic amongst its customers). If the firm is not able to quickly get back into the game with adequate capital, it is thrown out of exchange membership. The equity market stands alone in Indian finance with a vibrant creative destruction process: some firms die every year, and some new firms come about every year.

RBI transparency and what can be done about it

In March 2009, Dincer and Eichengreen wrote an NBER working paper where they found that RBI was one of the least transparent central banks in the world.

Today, on voxEU, there is an article describing a new paper on the subject: Central bank independence and transparency: Evolution and effectiveness by C.Crowe and E. E. Meade, European Journal of Political Economy which has a fresh attack on the problem of measurement of central bank transparency. They find that RBI transparency worsened between 1998 and 2006, and that the 2006 level was amongst the worst in the world.

The first task for the new RBI governor is to implement the article by Helene K. Poirson which explains nuts and bolts about what is wrong with RBI communication, and how this can be addressed. I think that a critical element of that is better data disclosure: RBI needs to start releasing currency trading data on a daily basis.

When the agency that sets the short-term interest rate is also selling bonds

Anup Roy has this fascinating article in Mint:

Ahead of the quarterly review of monetary policy on 29 July, the Reserve Bank of India (RBI) auctioned Rs6,000 crore worth 10-year bonds on Thursday at a cut-off yield of 9.0785%.

...

Following this, the benchmark 10-year bond yield, which was trading at 9.01% until late Thursday afternoon, rose to 9.09%.

While it is not exactly an indication of market expectations from the monetary policy to be unveiled next week, bond dealers say they are trying to figure out the underlying meaning of two important moves of the central bank involving the auction.

For the first time, buyers of govt bonds had to bid without being aware of the latest inflation figures

RBI was originally slated to auction a 14-19 year bond but, later pared the maturity to 10 years. It has also used the uniform pricing method to auction the bonds after a gap of two years. Normally, government bond auctions follow the practice of multiple pricing.

This is also for the first time the buyers of government bonds, such as banks, insurance firms and primary dealers, had to make bids without being aware of the latest inflation figures.

...

One can interpret the paring of maturity of the bond as a signal from the banking regulator on the interest rate outlook. The rates will go up and it does not want financial intermediaries to lock themselves in for longer period as they will have to take the hit and make good the depreciation in value of bonds, said a dealer who didnt want to be named.

J. Moses Harding, head of the wholesale banking group at IndusInd Bank Ltd, also echoed the same view. Probably RBI is taking a view that interest rates are on the higher side, he said. So, why lock in for a longer period and pay more?

Some other dealers say the central bank shifted to a 10-year bond for the auction because of the low demand for longer term papers. Insurance companies that generally buy longer term papers, dont have much surpluses to buy longer term bonds, said Harihar Krishnamurthy, head of treasury of Development Credit Bank Ltd.

According to Joydeep Sen, vice-president, advisory desk of BNP Paribas, the auction of a 10-year paper instead of a longer term bond, is possibly to reduce the borrowing cost of the government as longer bonds command higher interest rates.

Possibly the government is trying to fine-tune its borrowing cost and it also wants to issue more liquid papers to boost the sentiment, Sen said.

Under uniform price method, all successful bidders receive the bonds at the cut-off prices irrespective of their biddings while in multiple price auctions, bidders receive bonds at their bidding prices as long as they are within the range of the cut-off price.

According to bond dealers, the uniform pricing was adopted to shield investors from higher volatility.

Probably RBI wanted a better pricing. Uniform pricing normally ends up at slightly more bullish than multiple pricing. You dont have the bidding risk or winners curse there, said A. Prasanna, vice-president, ICICI Securities Primary Dealership Ltd, a firm that buys and sells bonds.

A winners curse happens when a bidder overpays to buy bonds in multiple auctions. In case the prices go down, the aggressive bidders lose the most.

The government wants to raise Rs96,000 in bonds in the first half of the financial year. So far it has raised some Rs66,000 crore.

A NewsWire18 story of 22 July quoted an unnamed finance ministry official as saying that the bidding was done on Thursday to beat release of inflation data.

The market is full of questions and speculation. Why was the date of auction not changed though inflation data release changed? Why was a 10-year bond issued instead of a 14-19 year bond as originally slated? Why was uniform pricing introduced? Since RBI is both central banker and investment banker to the government, do these decisions of RBI reflect knowledge of the future course of monetary policy being utilised to lower the cost of financing for the government?

You might be inclined to dismiss this newspaper article as reporting market gossip. But this kind of market gossip is a measure of the thinking of the market.

When the market is faced with these kinds of uncertainty, it ultimately damages the interests of government as an issuer. A customer is willing to pay more for a bond (i.e. a lower interest rate) when he is confident about what is going on. When a customer is confused (through lack of transparency) or worried that he's being had (through conflicts of interest), bond prices will be lower (i.e. the cost of financing will be higher) so that the customer is given a discount in return for the greater risk that he is facing.

This episode vividly illustrates the consequences of the mistakes in India's financial architecture. The investment banking function for the government should not be placed with the central bank. Markets must feel that the investment banker has not the slightest links to monetary policy. The investment banker must not have an inside track on knowing the future course of interest rates. When markets feel confident about this, when they know that the investment banker to the government is not playing games with them, markets will supply financing to the government at a lower cost.

There are exactly three ways to organise the debt management function:

1. To use the architecture of the RBI Act of 1934 (conflict of interest between monetary policy and investment banking), and then suffer from higher cost of financing for government as a consequence.
2. To use the architecture of the RBI Act of 1934 (conflict of interest between monetary policy and investment banking), but forcibly obtain low cost financing for government through financial repression -- i.e. forcing financial firms to buy government bonds.
3. To separate out the investment banking function into an independent Debt Management Office (DMO), which has no links with the central bank. This will reassure markets that the moves of the DMO do not reflect game-playing by the DMO based on a knowledge of the future direction of interest rates. This has been recommended by several expert committees, including the Mistry committee.

This is an example of what a proper institutional architecture is all about. In a mature market economy, the various functions of the State are placed in agencies without conflicts of interest. Each agency has a clear role and clear accountability, and operates with full transparency. In India, our landscape is filled with obsolete laws and obsolete architecture. Becoming a mature market economy is critically about making changes to the financial architecture.

We are growing up

The CMIE Cospi index is a total returns index of all firms in India where trading took place on atleast 66% of the days in the last six months. This is a weak liquidity filter. For all practical purposes, this is the Indian equity market.

The index level has fallen quite dramatically in recent months: from above 3200 to below 2000:

The bulk of this fall lies in a reduction in (trailing) P/E which dropped from 35 to 20. The world looks more gloomy as compared with what we saw in December 2007:

The market capitalisation of the index was above Rs.70 trillion; it's now below Rs.45 trillion:

The Economic Times would scream in their headlines that over Rs.25 trillion of wealth has been destroyed. But even at Rs.45 trillion, the stock market remains the most important component of Indian finance. As I often like to say, Indian finance is a plane running on one engine: the equity market.

I interpret the number of firms in the CMIE Cospi index as a measure of how many management teams in India have access to modern finance. This is has dropped by roughly 100 firms:

I'm impressed at the equanimity with which the Indian economy has digested this decline in stock prices. In the past, these kinds of large movements would have generated a lot more turmoil.

On one hand, every time large price movements took place, there would be accusations of a scandal. It would be claimed that there is some fraudulent foundation of the change in prices. Or, there were fears of a payments crisis whenever prices moved. And, sadly, India had a long history of difficulties on the equity market, so some of these fears were justified. But with the equity market reforms from 1994 to 2001, these fears have largely subsided.

Under socialism, prices did not change, so fluctuations of prices are often bewildering to people who spent their formative years under socialism. People who do not understand how financial markets work are prone to jump to the conclusion that these fluctuations are somehow wrong. On one hand, this leads to claims that the markets are a gambling den. I have met some senior economists in India, who had occupied prominent positions in government, who believe that all capital inflows into India are based on fads and whims, that there is no information processing or forecasting or economic logic underlying capital flows.

This kind of ignorance leads to demands that the government should do something every time prices change. When prices go up, the government is pressured to try to bring down prices. When prices go down, the government is pressured to try to push prices up. In the socialist worldview, prices are fixed, so all price volatility is bad.

But we're getting better: The CMIE Cospi market capitalisation dropped by Rs.25 trillion and we're hearing none of this. As the years go by, people out there in the real world are grappling with India as a market economy, and particularly the younger generation is coming to terms with the idea that prices must change in response to news; that a market where prices do not change is actually rigged.

So the good news is that we're growing up. When we attain this state of equanimity with fluctuations of the exchange rate, I will write a blog post saying that we have grown up.

On the subject of finance as a shock absorber, see Finance and growth: When does credit really matter? by Fabrizio Coricelli which offers new arguments about the mechanisms through which financial sector development influences economic growth in developing countries.