Is China over-investing in high-speed rail?
Is China over-investing in high-speed rail?
It's too early to judge
Given the speed with which China has built their high-speed rail network, are there issues with too much investment?
Does China Railway have too much debt?
These are hard questions to answer definitively at this point in time.
For ultra long-term investments like rail, it is really difficult to say at the time of investment (or even decades later) whether it was a bad decision. This is because these types of assets may be around for centuries and take multiple decades to realize. With the pace of change these days, I can barely imagine what life will be like ten years from now much less the middle of the 22nd century. And building out a national high-speed rail system is about as long-term an investment as any economy can undertake.
The question also asks whether the Chinese rail system has taken on too much debt. This is a slightly easier question to answer because the forecast time frame is a bit shorter (measured in decades as opposed to centuries) but it is still very difficult to properly answer — in part because comprehensive financial information is not publicly available but also because it is still really early in the grand scheme of things. I cannot provide a definitive answer to this question either, but what I do want to do here is point out the difference between societal gain and financial gain because this is also relevant to the broader question.
Some historical perspective
First, let’s step back and think about an analogous situation from the 19th century. I am referring to the great railway boom in the United States starting in the late 1820s and continuing through the early 1900s.
The famous Baltimore and Ohio (B&O) railroad was the first major rail line built in the United States with passenger service starting in May 1830. Other lines (disproportionately concentrated in the more industrialized Northeast) began to surface. Interestingly, many early locomotives were imported from England but very quickly we Americans copied these designs and established our own domestic locomotive industry (sound familiar?).
Railroads proved to be much more effective and economical than the alternatives at the time — steamboats/barges on rivers and canals and horse-drawn carriage on land. Railroads were faster, more comfortable and more reliable than all of these methods and initial high costs were rapidly brought down by advances in locomotive technology.
By the eve of the Civil War, railways connected most major cities in the young United States. They proved to be a decisive factor during the Civil War, giving the North a logistics advantage. There is a famous painting of Sherman’s famous March to the Sea that depicts Union soldiers tearing out a rail line — showing how important rail was to the Southern war effort.
After the Civil War, Americans embarked on its great westward expansion to fulfill our Manifest Destiny. Much of this effort was aided by a massive post-war investment in new railways across the western half of the continent (as well as rebuilding rail lines in the broken South). The First Transcontinental Railroad was completed in 1869, making it possible to traverse the continent in six days — a journey that had up to that point taken multiple months.
But good times for the industry ground to halt in the 1870s.
The Panic of 1873 was a major global economic depression which ended rapid rail expansion in the United States. Many lines went bankrupt or were barely able to pay the interest on their bonds, and workers were laid off on a mass scale, with those still employed subject to large cuts in wages. This worsening situation for railroad workers led to strikes against many railroads, culminating in the Great Railroad Strike of 1877.
And it would continue to be tough for the next century. The industry was forced to “rationalize” (which basically means that a lot of folks went bankrupt) multiple times. The advent of the automobile era — and later the Federal Highway System and airplanes — meant viable alternatives for both passenger (e.g. buses and later planes) and freight traffic (e.g. long-haul trucking). The freight rail industry was bogged down by heavy regulation, making it even more difficult to compete with truckers. The industry consolidated multiple times — yet still proved to be an incredibly difficult place for investors to earn a decent financial return.
Things finally started to get better in the 1980s, at least for the freight business. The Staggers Rail Act de-regulated the industry and made it easier to compete with long-haul trucking. Further consolidation reduced the number of players to just a handful of players with nationwide networks. The OPEC oil crisis led to a sharp rise in oil and gasoline prices — the main cost input for truckers. Good old-fashioned productivity improvement [1] within the freight rail industry further improved its competitiveness. And when Warren Buffett started buying shares of Burlington Northern Santa Fe in 2006 — and ultimately the entire company in 2010 — this was a pretty all-clear signal for investors to put their money back into the industry.
The difference between financial and societal gains
Sitting here in 2017, I think we can all agree that railway investment in the 19th century was ultimately a good thing. Railway investment unified the continent and integrated (both strategically and economically) the West with the East. Investment in railway technology catalyzed American industrial development. Towns and cities grew up around railroad stops, including some of the largest cities in the country. The United States would not be what it is today without railroads.
But many, many investors lost money investing in railroads. There were hundreds if not thousands of railroad bankruptcies. The boom-and-bust nature of railroad investment led to multiple economic depressions, not just in the United States but globally as well (see Panic of 1893) — Downton Abbey fans may remember how the fictional Crawley family fortune was put at risk because of a poor railroad investment.
Many people often wrongly equate financial failure with societal failure. “This project is not delivering a financial return today … ergo it should never have been built” This is actually my biggest pet peeve when it comes to people drawing conclusions on China’s high-speed rail system based solely on the financial performance at an isolated point in time. Not only does it often take significant time for utilization to ramp-up with long-term investment projects, just because something was a financial failure does not mean that society as a whole did not benefit from the investment.
Let’s look at the First Transcontinental Railroad to illustrate this point. After completion in 1869, the railroad shortened what used to be a dangerous and difficult journey (by ship or by land) that took months to one that took less than a week. It cost an estimated $150 million to build — even with probably a significant amount of corruption-related cost inflation. Ticket prices to cross the Continent decreased from $1,000 to as low as $65. This means you could reach “societal breakeven” in as little as 160,000 trips [2]. I could not find traffic numbers for the Transcontinental Railroad but if you look at the population growth of California alone in the 1880s and 1890s (>30,000 per year) and assume that most of this was from immigration, you will probably come to the same conclusion I did — the returns from a societal perspective were enormous. We are not even taking into account the savings from freight, the strategic benefits of binding the nation together, or the fact that we are still using much of this infrastructure nearly 150 years later.
But despite the enormous economic and societal benefits, a mere three years after the completion of the First Transcontinental Railroad, the Union Pacific Corporation — one of the three major companies financing the buildout — was forced to file for bankruptcy. Now much of this was due to corruption (Crédit Mobilier of America scandal) but bankruptcies would become a big part of the railway industry for the next century. Often difficulties generating a a decent financial return had more to do with things like over-competition that made it difficult for investors to capture the full value of the service that they were being provided to. In other words, the benefit to society from rail did not go away (improved travel times, job creation for millions etc.), it was just hard for some investors to capture at times.
With this historical perspective in mind, let’s turn our attention to the question at hand.
First, a few important characteristics about China’s economy in the early 2000s (i.e. around the time it was starting to seriously consider high-speed rail investment) that you need to be aware of:
Slack labor capacity — hundreds of millions of relatively young people living in rural villages that were not engaged in the modern, formal economy in a productive capacity.
Land acquisition was straightforward — land largely owned by the State so it was easy to secure wide swathes of land between cities.
Relatively low quality of life expectations — citizens that tended to still care more about basic development needs (i.e. the “right to a full stomach”) than those that are higher up on Maslow’s hierarchy of needs like the “right to live in your hovel and block infrastructure construction”.
At the time, the conditions to build high-speed rail were pretty ideal. Labor cost was cheap. It was easy to acquire paths in straight lines between cities. There were economies of scale to undertaking a massive build-out — including negotiation leverage with foreign technology suppliers.
In the future, these conditions might not exist. Already today, China no longer has the same amount of slack labor capacity and per unit labor cost is higher. As its cities have grown, the cost to build in an already developed area becomes exponentially more difficult. As China’s citizens have gotten richer, quality of life expectations have risen and they have “discovered” NIMBY-ism — which only compounds the cost and difficulty of building.
So there are a lot of positive arguments to undertake this build-out today vs. the future when costs could rise tenfold.
That said, there are also some good reasons why doing everything at once could also result in negative effects for society.
Higher commodity prices — intensive infrastructure build-out (not limited to just railways) resulted in a spike in the usage of things like steel, copper and cement. China needs to import a lot of these commodities and correspondingly its import bill skyrocketed, resulting in higher build costs than if they had taken a less intensive path.
“Bridges to nowhere” — when you throw tons of money at infrastructure, the inevitable outcome is ending up with a lot of stuff that never gets used. At best, unused railway lines are a waste of resources that could have been better spent somewhere else. At worst, you actually need to spend a lot more money in the future just to get rid of it. There are numerous examples of these in China across many infrastructure related industries.
Corruption — when you throw a lot of money at infrastructure in an economy with relatively weak institutions, another inevitable outcome is that some of this money ends up lining the pockets of a privileged minority. This leads to greater wealth inequality and social discontent.
Societal costs of debt restructuring and bankruptcy — much of China’s high-speed rail build-out has been financed by debt. While almost 100% of this debt has been borrowed domestically (i.e. a “son owes his father” vs. “owe your neighbor” situation), too much debt still increases the risk of a future economic crisis and there are significant real-world costs associated with crises.
The answer to this question will ultimately depend on how positive the positive bullets are offset by how negative the negative bullets are:
Would easier land acquisition and lower labor costs have been offset by higher commodity prices if some of the build-out were delayed?
Would it have been better if the resources used in building high-speed rail networks were invested/consumed in other areas (e.g. real estate, foreign imports)?
Would the costs of corruption have been mitigated in the future when institutions are presumably stronger?
Would the rail network result in greater utilization if it had been built out more methodically and slowly?
Do the societal and financial profits from certain heavily used lines (e.g. Beijing to Shanghai) more than subsidize the societal and financial losses from lines that see less utilization?
How might future technologies (e.g. autonomous driving) impact the societal and financial returns of high-speed rail?
Are we factoring in societal gains that are not necessarily captured in the financial P&L like job creation for millions and lower pollution?
What are the second and third-order effects of building out a nationwide high-speed rail network?
As you can see, these questions are exceedingly difficult to answer, especially today when we are still in the middle of this build-out and much of it is hypothetical anyway. I think it will take many decades for us to really know the answer to this question — heck, five decades later I think we are still struggling to answer the question of whether we over-invested in the Interstate Highway system in the 1950s and 1960s.
But if I were forced to answer, the best I can do is to say that I do think early indications are positive overall, as I mention in another answer. [3]
Notes
[1] From Berkshire Hathaway’s 2015 Annual Letter, Warren Buffett wrote about the railroad industry:
In 1947, shortly after the end of World War II, the American workforce totaled 44 million. About 1.35 million workers were employed in the railroad industry. The revenue ton-miles of freight moved by Class I railroads that year totaled 655 billion.
By 2014, Class I railroads carried 1.85 trillion ton-miles, an increase of 182%, while employing only 187,000 workers, a reduction of 86% since 1947 …
… As a result of this staggering improvement in productivity, the inflation-adjusted price for moving a ton-mile of freight has fallen by 55% since 1947, a drop saving shippers about $90 billion annually in current dollars.
Another startling statistic: If it took as many people now to move freight as it did in 1947, we would need well over three million railroad workers to handle present volumes. (Of course, that level of employment would raise freight charges by a lot; consequently, nothing close to today’s volume would actually move.)
Our own BNSF was formed in 1995 by a merger between Burlington Northern and Santa Fe. In 1996, the merged company’s first full year of operation, 411 million ton-miles of freight were transported by 45,000 employees. Last year the comparable figures were 702 million ton-miles (plus 71%) and 47,000 employees (plus only 4%). That dramatic gain in productivity benefits both owners and shippers. Safety at BNSF has improved as well.
[2] $150 million cost divided by $935 in savings per passenger-trip. Clearly it wasn’t just the capital investment that needed to be paid back as railroads run significant operating expenses. But ongoing operating expenses were comprised largely of labor and the creation of millions of new jobs has clear societal benefits as well.
[3] I am basing this on a few simple numbers
Total investment in Chinese high-speed rail system has been around $300 billion as of early 2017.
In 2016, there were 1.4 billion high-speed rail trips taken.
If passengers saved an average of about one hour per trip, that is 1.4 billion hours saved per year and this number is expected to increase significantly in the future. Average urban wages in China are about $5/hour and will likely double in five years.
That means around $7 billion in return just from time savings — rising to something like $30 billion in five years. This is just time savings, one small part of the overall gain for society.
It ignores the millions of jobs created, the higher tax base to fund government services, the second-order effects of being able to conduct business faster, the positive effects on pollution and many other positive externalities.
And these are returns that are likely to continue to increase at a rapid pace for decades to come as society becomes even more productive and wealthy.
This was originally published on Quora in June 2017.