SEARCH

TSI: `A good measure of how blood flows in an economy' Mr Kajal Lahiri, Professor of Economics, State University of New York

print   ·  

Pratap Ravindran

ON January 29 last year, the US Secretary of Transportation, Mr Norman Y. Mineta, ringing the opening bell at the New York Stock Exchange, announced the rollout of the first-ever Transportation Services Index (TSI), which he called the "Dow Jones of the transportation sector" and "an exciting, new way to track the transportation industry's huge impact on America's growing economy."

Mr Mineta went on to elaborate: "The index focuses on two very important transportation components the movement of freight by land, water, and air, and the movement of people on the ground and in the skies. Month by month, the index will measure changes in the amount of freight being moved by a range of "for hire" industries representing rail, truck, air, and inland waterway transportation sectors.

"It also tracks monthly changes in rail, bus, and air passenger traffic. This is important because the "for-hire" transportation industries, like other service industries, have not traditionally been well represented in economic indicators. The index fills that huge void and gives us a solid reading on where the economy is headed... ."

The rollout of the index received extensive coverage in the US media but none at all in the Indian financial press, which was preoccupied with the Q3 results of Reliance Industries Ltd, a record profit reported by Steel Authority of India Ltd (SAIL), the entry of the Sensex into a "correction mode" and other equally momentous subjects at that time.

This would have been entirely in keeping with the nature of things if not for the fact that the creator of the index is an Indian.

Ironically, the domestic media, which usually becomes hysterical when reporting any accomplishment or even endeavour by an Indian, however small, dropped the ball when it came to reporting the work of an academic from India who had made an extremely significant contribution to economics.

Mr Kajal Lahiri, a professor of economics at the State University of New York at Albany and Director of its Econometric Research Institute, along with a doctoral student, Mr Vincent Yao, had been working since October 2001, on developing ways of tracking activity along the American highways, railroads, flight paths, waterways and pipelines in order to create an index which, by interpreting the movement of people and goods, would help economists figure out where a given economy is in the business cycle and the direction it is likely to take.

In due course, the research project, sponsored by the US Bureau of Transportation Statistics, yielded an economic indicator as valuable as the Gross Domestic Product (GDP) and the Consumer Price Index (CPI) and, on March 10, 2004, the first ever report on the impact of transportation on the economy as measured by the TSI comprising eight components, ranging from trucking tonnage through mass-transit ridership to petroleum pipeline transport was released in the US.

BusinessWeek, among other prestigious publications, was gung-ho about the index. It quoted Mr Lahiri as saying that the index was quite sensitive to shifts in economic growth because nearly all businesses rely on the transport sector the importance of which had increased in recent years because of the widespread use of just-in-time inventories and production schedules.

Validating Mr Lahiri's description of the TSI as a good measure of "how the blood is flowing" through the economy, the periodical noted: "Indeed, analysis by BusinessWeek shows that yearly growth in the TSI and the Federal Reserve's market-moving industrial production index are highly correlated."

It added: "The index also foreshadows economic downturns, says Mr Lahiri, because trucking and rail lines quickly reflect a drop in goods demand. Plus, passenger travel typically slows when business activity weakens. Using the professor's original data going back to 1980, the index has signalled economic contractions a year before they occurred, on average. The index fell 2.5 per cent in the year before the last recession, which began in March 2001. In addition, the index provides another sorely needed measure of the service sector. Services constitute about two-thirds of the economy, yet few government reports cover them."

The issue now is whether the index can be constructed and used in India. The country needs a new economic indicator badly the forthcoming Budget will be packed with references to GDP, now acknowledged by economists to be a fatally flawed indicator in that it is a mere aggregate of expenditures which does not provide any meaningful insight into the economic welfare of society. On the contrary, GDP perversely rises whenever an economy is hit by, say, a natural calamity such as the recent tsunami, because of the consequent increase in spending much to the delight of the Finance Minister of the moment who invariably makes full use of this aberration to gull the public.

If the TSI is put in place in India, won't its accuracy be impacted by the time-lag that will inevitably be involved in the provision of data by the government-run entities the railways, for instance as also the fact that the bulk of the truck operators in India continue to be in the informal sector and do not maintain accurate records because of tax and related considerations?

And then, again, won't the TSI suffer from the same infirmity as the GDP, going up whenever there is a large-scale disaster?

This correspondent put these questions to Mr Lahiri, and the following are his answers:

On dealing with the availability of data:

Your concern provokes a strong reaction from me as an educator. Good research is always difficult, whether you are in India, Africa or America. Different places have different sets of problems. India is not special in this regard. A smart researcher identifies them and tries to solve them to the best of his or her ability.

In India you refer to certain genuine problems that exist in the US too, but may be to a lesser extent. On the other hand, the US faces a different set of obstacles (for example, more stringent legal issues, privacy laws, etc). I agree that with monthly indicators like the TSI, the timing issue is very serious and critical.

For quarterly series (like the GDP) it is less of a problem. There have been many research efforts that indirectly measure the activities of the informal sector. For indicator analysis, it is not important to measure everything so far as the series tracks the up and down movement of the underlying sector.

One advantage of India in this context is the relative importance of the railways (like in the US 50 years ago) that must have good data on goods and passenger movements, even on a weekly basis. Realise that once goods are moved by major carriers like the railways and leading trucking lines, they have to go eventually somewhere to the consumers.

So all the movements by everyone may not be necessary to gauge the movements in the transportation sector once the major movers have been counted for.

A good, insistent researcher can procure major data provided one gets the blessings from the top (like the Central Ministries) and is supported by some major professor in research or educational institutions. In India I have seen many truly committed young researchers who can sweat out such difficulties, provided he/she is persistent and has the support from the top.

Here my research group could do it because the Secretary of Transportation wanted the series to be developed. Regarding lags, the monthly series, like consumer sentiment, industrial production, TSI, etc., appear five weeks after the end of the month. This lag is the shortest. In India you do the best you can do. One can also construct a provisional series as soon as the major sectors are covered.

For quarterly data, it is less of an issue. I believe that railway movements in India can be measured promptly. After all they've been running on a super computer for a long time now. Trucking could be difficult, but as long as there is a regulatory agency, which oversees their activity, one can get the information with friendly pressure.

Since one is interested in the aggregate activity, information from major individual companies can be obtained provided the privacy and security is assured. The trust has to be developed. Going through private trucking organisations can ensure this. That is what I did here by going through the American Trucking Association, which is a private trade organisation. After all, individual trucking companies benefit when they get the aggregate information.

These are good research questions that should be explored in India. I agree that it will have different sets of issues to solve. There are no easy answers. But in India, a series like TSI can be constructed by hard work and determination.

On the possibility of the TSI suffering from the same infirmity as the notion of GDP:

The issue is important and has been considered by business cycle analysts. Erratic movements (strikes, natural disasters, etc.) and other movements (like seasonal effects) in an economic time-series (that an analyst may not be interested in) can be smoothed out by appropriate smoothing algorithms.

Over the last 50 years, many such algorithms have been developed with success. The Hodrick-Prescott (HP) algorithm is one of them. Our first paper, which appeared in the Journal of Transportation and Statistics (JTS), published by the Bureau of Transportation Statistics of the US Department of Transportation, explains the logic and the intuition behind the Index.

In our case, the classic recent example is the effect of the 9/11 incident on air passenger traffic.

Incidentally, the GDP is a quarterly series, and for monthly series this problem is more severe. The effects of known incidents can be netted out without much of a problem. This is true for Indian GDP too.

The only problem with some of these smoothing algorithms is that they may artificially push the peaks and troughs of the series to the right a bit.

However, HP and a few others do not. Thus the infirmity you refer to can easily be avoided and explained to the lay public.

(This article was published in the Business Line print edition dated February 23, 2005)
XThese are links to The Hindu Business Line suggested by Outbrain, which may or may not be relevant to the other content on this page. You can read Outbrain's privacy and cookie policy here.

O
P
E
N

close

Recent Article in OPINION

WTO must prevail over splinter groups

Trade between members of a group affects the rest of the world. Hence the importance of multilateralism »

Comments to: web.businessline@thehindu.co.in. Copyright © 2014, The Hindu Business Line.