Thursday, September 19, 2013

Why Spending RM160 Billion On Rail Isn’t Completely Bonkers

It’s last week’s news, but still relevant (excerpt):

Najib: An additional RM160bil to be invested in rail projects by 2020

KUALA LUMPUR: The Government is expected to spend an estimated RM160bil more on rail-related projects until 2020, said Datuk Seri Najib Tun Razak.

The Prime Minister said the railroad industry had seen "massive expansion" in Asia and was becoming an increasingly significant mode in Malaysia's efforts to improve public transportation...

...Adding that the Government had invested over RM50bil since 1990, he stressed the need to train skilled and specialised manpower to fill hundreds of thousands of jobs generated from rail projects...

...Once fully operational, he said the MRT would cover 156km and serve up to two million trips per day, a huge jump from the 500,000 daily trips on the current urban rail system.

Najib added that another project in the pipelines was the 330km-long Kuala Lumpur-Singapore high speed rail link, estimated to be completed by 2020, which would cut land travelling time between the two countries to just 90 minutes...

When the MRT and other public works under the Economic Transformation Programme were mooted, there were some complaints that these were no more than property plays (TRX especially comes to mind). These concerns would especially be valid in the current environment, with fiscal consolidation in the wind (not that I agree with it, but that’s another story).

RM160 billion is a pretty significant amount to finance, equivalent to 80% of the government’s operating budget this year. Can we afford to do this?

But there’s long term value in “property plays”, if the focus is purely on transportation. Two new NBER papers illustrate some of the mechanisms, the first focusing on the impact of transportation infrastructure in opening up economic opportunities (with an emphasis on agriculture), and the other focusing on the impact on the real exchange rate (abstract):

Railroads and American Economic Growth: A “Market Access” Approach
Dave Donaldson, Richard Hornbeck

This paper examines the historical impact of railroads on the American economy. Expansion of the railroad network may have affected all counties directly or indirectly – an econometric challenge that arises in many empirical settings. However, the total impact on each county is captured by changes in that county's “market access,” a reduced-form expression derived from general equilibrium trade theory. We measure counties' market access by constructing a network database of railroads and waterways and calculating lowest-cost county-to-county freight routes. As the railroad network expanded from 1870 to 1890, changes in market access were capitalized into county agricultural land values with an estimated elasticity of 1.1. County-level declines in market access associated with removing all railroads in 1890 are estimated to decrease the total value of US agricultural land by 64%. Feasible extensions to internal waterways or improvements in country roads would have mitigated 13% or 20% of the losses from removing railroads.

Translation: transportation fosters development. But, we have to be clear here about the purpose of investing in such infrastructure. The paper primarily looks at the impact of rail transportation for trade, as much as land values. You’re not going to get a return if you build a railroad serving rural areas only; it has to be part of a larger network of transportation. The KL-SG link makes sense, as does the MRT. So too would a pan-Borneo railway, if it helps open the hinterland to trade opportunities, or increases trade with the Indonesian south.

Roads and the Real Exchange Rate
Qingyuan Du, Shang-Jin Wei, Peichu Xie

This paper studies the effect of transport infrastructure on the real exchange rate (RER) and reaches two relatively strong conclusions. First, while the list of robust determinants of the RER is not long, transport infrastructure belongs to that list. Many other potential determinants proposed in the literature, such as net foreign asset position or terms of trade, turn out to be not robust. Second, in terms of economic significance, the infrastructure effect follows closely the well-known Balassa-Samuelson effect and is one of the most important explanatory variables for RER movements, especially in developing countries.

This paper in turn looks at the effect on the exchange rate of investment in transportation. The main effect would be to reduce freight costs, and induces a depreciation of the real exchange rate i.e. it makes you more competitive. Again however, we need to look at the cost-benefit of any such investment, given that the primary effect that you want is to reduce the costs of making and selling tradable goods.

One other key point here is that rural development is predicated on goods roads. I remember a time when travelling from KL to my Dad’s kampung was an 8-12 hour ordeal. Now it takes 3 hours, and in much greater comfort. Roads make the world a smaller, easier place to work and live in.

So, does all this justify spending RM160 billion on rail? I’d give it a qualified yes.

Technical Notes

Dave Donaldson, Richard Hornbeck, "Railroads and American Economic Growth: A “Market Access” Approach", NBER Working Paper No. 19213, July 2013

Qingyuan Du, Shang-Jin Wei, Peichu Xie, "Roads and the Real Exchange Rate", NBER Working Paper No. 19291, August 2013


  1. Inflated cost...opaque process.Budget and feasibility not tabled and approved by Parliament thus project is off budget and in stealth mode.
    Its not railways thats in issue...only the Great Train Robbery.

  2. Inflated cost...opaque process.Budget and feasibility not tabled and approved by Parliament thus project is off budget and in stealth mode.
    Its not railways thats in issue...only the Great Train Robbery.

  3. ... which comes to 22B a year ... fantastic!

    1. @anon 6.35

      The development budget, which includes investment in roads, schools, utilities and yes, rail, has exceeded RM40b every year for the last six years.