THE Land Transport Authority's (LTA) decision to run a one-year pilot that gives free train rides to commuters travelling to the city on weekdays before 7.45am has captured the public's attention.
There are some who wonder if the move will simply create a new peak period. Others question whether taxpayers should pay for the fares of some commuters, especially since the Government has long eschewed operating subsidies for public transport.
Yet others think that the experiment favours those commuting to the city and discriminates against those who commute by bus and those who work outside the city.
Some of these concerns are valid. Nonetheless, it will be unfortunate if these criticisms detract from the usefulness of the pilot project. This pilot is valuable in at least two respects.
First, it reflects the Government's willingness to harness the insights of behavioural economics to solve what seems to be an intractable problem. Second, it demonstrates the Government's willingness to question its longstanding aversion against providing anything for free.
The year-long pilot offers policymakers an opportunity to test if free train rides during off-peak hours are sufficient to produce an enduring change in some people's commuting habits.
According to behavioural economist Dan Ariely, zero is not just a low price; it is a source of irrational excitement. People strongly prefer free items even if a better deal is available at nominal cost. For instance, shoppers prefer getting something free (say a "buy two, get one free" promotion) to getting something cheaper (a "buy three, get 40 per cent off" promotion).
In a well-known experiment, economists Kristina Shampiner, Nina Mazar and Dan Ariely found that when expensive Lindt truffles were sold at 15 cents and inexpensive Hershey's Kisses at one cent, 73 per cent of their experimental subjects chose the Lindt truffles.
However, when the experimenters decreased the price by one cent apiece, so that Lindt truffles cost 14 cents and Hershey's Kisses were free, 69 per cent chose the latter.
Standard economics does not explain this "power of free". Rational cost-benefit analysis predicts that decreasing the price of both chocolates by the same amount should result in no change in the relative demand for both chocolates.
By providing free train rides, the LTA hopes to harness the power of free to shift demand from peak to off-peak travel. Congestion is a 10 to 20 per cent phenomenon; transport planners do not need to shift a majority of commuters to off-peak hours. As long as 10 to 20 per cent do, that is sufficient to alleviate congestion and improve the commuting experience for all.
But shifting 10 to 20 per cent of commuters may not be as easy as the power of free suggests.
For many of us, the only times when we make a conscious and deliberate decision about when to begin our commute are during the first few days on a job or on rare occasions - such as when we have an unusually early meeting.
But outside of these occasions, once our commuting decision has been made, we tend to rely on routines that unfold automatically every day. This power of habit suggests that unless something that disrupts our routine occurs, we are unlikely to change.
Whether the power of free would triumph over the power of habits for at least 10 to 20 per cent of commuters is an empirical question. The pilot experiment will reveal to us the extent to which this is true.
Second, the pilot is valuable in that it reflects a welcome willingness on the part of the Government to question its longstanding aversion against providing anything for free. At the heart of this philosophy is the assumption that anything which is provided freely will be undervalued and over-consumed.
By making train rides free, even if temporarily, the Government has signalled that it is not afraid of questioning its own ideology to solve a pressing problem.
In this instance, an anti-free ideology is questionable. Is it really the case that providing free train rides at particular times would cause a rush for them? Would there really be commuters taking train rides into the city before 7.45am just for the fun of it?
But if the Government realises that there are instances where the risks of over-consumption and moral hazard are overstated, this raises the question of whether there are other policy areas where its fears of moral hazard need to be tested against evidence.
The Government could, for instance, apply this empiricism to areas such as health and long-term care.
The usefulness of an experiment also depends on the existence of a robust measurement and accountability framework.
There are several ways in which public sector agencies may fall short on this score. For instance, bureaucrats may fear the consequences of a failed pilot and may rig the experiment such that they increase its likelihood of success, but in ways that cannot be sustained after the pilot period has ended.
Alternatively, fearing a public backlash should the experiment fail, bureaucrats may be less than transparent with its results, or they may refuse to subject themselves to the rigorous standards of research.
To address this risk and to increase public confidence in this regard, the LTA should consider commissioning independent researchers from our local universities when analysing this pilot. These results should also be made publicly and freely available.
Alisha Gill is a researcher at the Lee Kuan Yew School of Public Policy. Donald Low is senior fellow and assistant dean (research centres) at the school.
This story was first published in The Straits Times on April 20, 2013
To subscribe to The Straits Times, please go to http://www.sphsubscription.com.sg/eshop/