Public policy challenges - case of Bangladesh's electricity sector

FT has an excellent long read in the context of Bangladesh that highlights several policy challenges with the implementation of industrial policy and the green energy transition. 

When Sheikh Hasina came to power in 2009 in Bangladesh, the country was suffering from electricity blackouts. She responded with policies that encouraged new fossil fuel plants like generous capacity payments and protection from legal challenges and prosecution, and eased imports of LNG.

Electricity generation capacity soared from 5.5GW in 2009 to 23GW currently, according to the Institute for Energy Economics and Financial Analysis, a think-tank. As a result, Bangladesh’s entire population of 170mn now has access to electricity, compared with less than half in 2009. Prioritising gas helped limit the country’s dependence on coal. While coal imports have also grown, it accounts for 12 per cent of the country’s generation capacity compared to 50 per cent for gas, according to the BloombergNEF, a commodity research service. In 2021, Sheikh Hasina cancelled plans to build 10 new coal-fired power plants. 

However, the reliance on imported natural gas and capacity expansion has now created a set of problems. On the one hand, is the problem of high gas prices and large capacity addition,

Yet Bangladesh now faces problems of a different nature. Electricity demand did not keep up with the building frenzy, and total capacity now exceeds demand by as much as 50 per cent. The lack of investment in new domestic hydrocarbon exploration also means the country’s domestic gas reserves are running low. That has left it at the mercy of global LNG prices. Unlike piped gas, LNG can be transported anywhere there are regasification facilities and cargoes tend to gravitate to the markets where prices are highest. Following Russia’s invasion of Ukraine, that market was Europe — prices rose so high that traders were willing to pay penalty clauses to Asian countries in return for diverting their cargoes westward. Fuel shortages and a surging energy import bill pushed Bangladesh into one of its worst crises in years, with rolling blackouts and painful inflation. Foreign currency reserves have fallen about 20 per cent this year, according to rating agency Fitch, and the country has taken out a multibillion-dollar IMF loan to steady its economy.

On the other hand, there’s the issue of generous fiscal commitments made by the government to incentivize these investments 

This means that even as Bangladesh struggles to afford fuel, more money goes towards capacity payments for projects that have in some cases been idle for more than a year. Bangladesh’s power sector subsidy burden jumped to Tk297bn (about $2.7bn) in 2022, the IEEFA said, 150 percent higher than a year earlier. The Summit Group has been a leading beneficiary of Bangladesh’s energy policies. Along with its upcoming plant in Meghnaghat, it is building several new LNG projects in addition to about 15 existing power plants. Of the Tk1tn ($9bn) of capacity payments by Sheikh Hasina’s government since it came to power, Summit has in recent years been the largest private beneficiary. Muhammed Aziz Khan, Summit’s founder, argues that Bangladesh has no choice but to import more LNG if it wants to avoid burning coal… to critics, the continued gas buildout shows that the interests of these powerful and politically connected conglomerates have long since overtaken considerations about energy security.

Rashed al Mahmud Titumir, an economist at the University of Dhaka, summarises the problem

Overcapacity, high retail prices, fuel shortages and now a struggle to pay energy bills — while subsidising an oligarchic clientele.

The private plants generate power at about Tk10 per kilowatt hour, roughly double the cost of that generated by state-owned companies. 

The obvious option of renewable generation faces several problems

Analysts say land is in short supply in the densely populated country, while financial incentives continue to make fossil fuel projects more attractive to the private sector, resulting in a shortage of funding for renewables… Laurent Ruseckas, executive director for gas at S&P Global Commodities says that the lack of local supply chains for the construction of renewable capacity means that it’s not “seen as plausible to skip gas” in countries like Bangladesh. “You still need gas in those markets. The challenge for those countries is how to fit it into the energy system and make prices work.”

Some observations:

1. Even with all its failings, Bangladesh’s success in catalysing private electricity generation is a very impressive achievement for a lower middle-income country. For perspective, look at Pakistan or countries in Africa. 

2. Bangladesh’s case is also a good example that illustrates the challenges associated with energy transition. Bangladesh is better off in so far as it has placed its bets on gas based generation and not coal. But having yoked itself to the natural gas, and that too recently, it’ll be difficult, almost impossible, for Bangladesh to transition to renewables in any meaningful manner for the foreseeable future. 

3. This is a classic example of the need for industrial policy to be dynamic. The decision in 2009 to provide generous incentives to attract private investments into generation was the right one. It can be argued that the incentives were too generous, and not structured to be demand-driven. Further, once investments started flowing in and projects got commissioned, the government should have started to phase down the incentives for future projects. 

4. But calibrating and phasing down industrial policy incentives is easier said than done. And that’s perhaps the most compelling critique of industrial policy. European countries suffered in the noughties with its generous feed-in-tariff incentives for wind and solar generation. The Production Linked Incentive (PLI) scheme in India is an example of an industrial policy that needs constant monitoring to tweak based on emerging trends. 

5. But for countries like Bangladesh, it’s difficult to imagine an alternative pathway to rapidly meet its electrification requirements. This is the Big Push required to catalyse investments and markets at such scale. Such industrial policy with sufficient enough conscious policy slack is required to induce the level of private investment required. In purely economic terms, this policy slack is inefficiency and waste. A similar example of conscious slack is perhaps that of condoning a few large business groups with the risk appetite (and political connections) to become dominant in infrastructure and related sectors. But then these large groups foster crony capitalism and end up perpetuating the industrial policy.

6. In developing countries, where markets are emerging, there’s a strong case for the public sector to assume leadership and remain the largest investor till markets become mature enough. There’s also the case for governments to remain invested so as to keep the markets honest. In India, the presence of state gencos, and more importantly NTPC, have been critical, but less acknowledged, factors in the development of a reasonably equitable and efficient market in electricity generation. 

7. Finally, this example yet again illustrates the elephant in the room in the power sector, the viability of distribution. The sector’s sustainability is critically dependent on the ability of distribution companies to recover at least the cost of supply from households. In its absence, all reforms and private capital structuring is of no use. Governments cannot absorb losses forever and there are also limits to how much losses can be subsidised by the government each year. There are also limits to how much cross-subsidy can be borne by industry and commercial sectors. This is a political economy challenge that even most Indian states have not been able to overcome. To this effect, electricity is midway between a private good and public good - it’s de facto non-excludable (and tariff pricing to recover costs is also not possible)!

Thoughts on Affordable Housing V

I have blogged on several occasions that the only solution to making housing affordable is to expand supply. This supply expansion has to combine both market and public housing supply. The former requires combining multiple policy levers and staying the course with those policies for a long enough time, and the latter requires large fiscal spending. The earlier posts in this series are here, here, here, and here. 

John Burn-Murdoch in FT makes the same argument about increasing supply in general, and writes that the trickle-down effects work in the affordable housing market in developed countries.

But recent studies from the US, Sweden and Finland all demonstrate that although most people who move directly into new unsubsidised housing may come from the top half of earners, the chain of moves triggered by their purchase frees up housing in the same cities for people on lower incomes. The US study found that building 100 new market-rate dwellings ultimately leads to up to 70 people moving out of below-median income neighbourhoods, and up to 40 moving out of the poorest fifth. Those numbers don’t budge even if the new housing is priced towards the top end of the market.

He also points to recent research questioning the gentrification argument on squeezing affording housing supply,

Another argument is that building market-rate housing in a lower-income area leads to gentrification, with higher earners moving into a lower-income area and displacing the incumbents. But the latest research from Britain and the US shows that there is typically little, if any, outward displacement of incumbents. It is the incomers who have been displaced, priced out of wealthier areas by supply constraints. In other words, even if you think it’s inherently bad if high earners move into poorer neighbourhoods, the answer is to build more market-rate housing for those higher earners.

A new paper by Geetika Nagpal and Sahil Gandhi explicitly addressed the question of whether the relaxation of zoning regulations increases affordable housing or simply triggers the building of new luxury units by examining data on the effects of a 2018 FAR deregulation in Mumbai. 

Leveraging granular panel data and exploiting variation in time and space, we find that the reform increased housing supply in treated areas by 28%, implying an elasticity of housing supply to the FAR of 1.59. The FAR relaxation increases the scale of development, resulting in higher investment in shared amenity space within the building. This increased public good provision facilitates an 18% decline in unit sizes, leading to a 29% decrease in apartment prices that allows lower-income households to access housing. We develop a structural model of housing supply and demand that incorporates the provision of amenity floorspace and shows that after the relaxation, average home buyer incomes are 3.18% lower. We use the estimated model to show that a further 5% rule-based relaxation would amplify the scale economies and increase the affordability gains from deregulation by 1.7%. Taken together, our results show that concentrating FAR relaxation can improve affordability.

In this context, I have blogged that affordable housing policy instruments should not only target supply but also specifically target the genuinely affordable housing supply. I argued this, especially in the context of large developing countries like India, where the baseline market conditions are very different from those in developed countries. In these countries, the demand-supply mismatch covers the majority of  the population and their affordability gaps are much higher than in developed country cities.

The housing market in developing country cities consists of at least four segments. The first is the richest sliver, who can afford housing in the main parts of the major cities. Second are the higher income and upper middle class whose affordability is confined to outside the main core of the city, third are the middle and lower middle class who have incomes but enough to buy only at the margins of even the suburbs and face the genuine affordability problem, and finally those who cannot afford any market supplied housing and thereby end up in slums and squatter settlements. 

The third and fourth categories make up the vast majority of the population, certainly over 80%. The affordability gap for the third group is very high, in orders of magnitude of their current affordability. As an illustration, assuming an annual income of Rs 6 lakh for these households and affordability of ten times annual income, the supply of Rs 60 lakh units will however be only at the margins of even the suburbs, if at all. The fourth group will certainly require some form of heavily subsidised housing or public housing in rental mode. 

In this context, what's of relevance is not the unit size, but the price. Even within the price, it's the unit price declines that are of critical importance. And even when there are reductions, they have to be high enough to make any meaningful dent in the problem. 

This is because the supply of market rate units will mostly serve customers with higher incomes and this market itself is large enough to absorb most of the incremental supply for long years. The trickle-down effect does not materialise in any meaningful manner in a finite time period.

There are several issues with drawing the general conclusion that FAR deregulation leads to an immediate and significant enough impact on affordable housing supply from the paper. For one, the paper does not discuss what constitutes affordability and the aforementioned market segments. I'm not sure about the meaning of the 29% decrease in apartment prices, since the relevant metric is either the price for comparably sized units or the unit price of housing (per sq ft price). It's impossible to have anything close to such a decline in apartment prices on the same or similar metrics. In absolute terms, given the existing affordability gap, an affordability gain of 1.7% from deregulation would be insignificant. Further, the 28% increase in housing supply in treated areas corresponds to just a 0.7% increase in aggregate housing stock added to the city each year, and tiny addition compared to the requirement. Or the 3.18% average income of the new buyers would itself be so small and confined to within the top income quintile. There are also important limitations to the data sources used in the paper.  

Finally, the deregulation was itself limited to sites abutting roads that are more than 12 m (or 40 ft), which invariably rules out those localities that predominantly supply housing for those in the third and fourth categories. This also raises the point that a meaningful dent in housing affordability in large cities requires large deregulation and renewal in slums and lower-income localities. This is also important since the majority of land available in major cities would be those with less than 40 ft abutting roads. At the least, the land values in those places within large cities with more than 40 ft abutting roads would be much higher than in those with smaller abutting roads. 

In the net, I'm inclined to conclude that such limited deregulation would end up only slightly lowering the prices, increasing the affordability, and improving the quality of housing supply aimed for the second segment, but would have only negligible impact on the third and fourth market segments. 

But having said that, the problem is so complex and intractable that any meaningful effort to address it would require the adoption of all possible measures to increase supply. And there's nothing more important on the regulatory side than easing zoning restrictions, easing them significantly and especially in terms of much higher FAR norms. The point I'm trying to make is that merely deregulating and deregulating even aggressively with a much higher FAR will only take you so far in addressing the affordability gap. 

On this point, Burn-Murdoch also points to the examples of Auckland and Minneapolis, both of which relaxed zoning regulations with impressive results in increased supply, and that too in a relatively short time period. 

In November 2016, large areas of New Zealand’s largest city, Auckland, were rezoned to allow for higher-density building. The results were twofold: a boom in construction of multi-unit housing — predominantly at market rates — and the flattening off of rents in the city in real terms. On the eve of upzoning, median rents were 25 per cent higher in Auckland than the capital Wellington. Six years later, nominal rents had grown by an average of 3 per cent a year in the former and 7 in the latter, putting the two neck and neck. Adjusted for inflation, renting in Auckland is now no more expensive than it was in 2016, compared with a 25 per cent rise in Wellington.

In the American Midwest, where Minneapolis has been building more housing than any other large city in the region for years, and has abolished zones that limited construction to single-family housing. Adjusted for local earnings, average rents in the city are down more than 20 per cent since 2017, while rising in the five other similarly large and growing cities.

He also points to a new paper by James Gleeson that points to the importance of increasing the housing supply in making housing affordable.

The latest issue of Works in Progress has an essay by Eleanor West and Marko Garlick on zoning reforms undertaken by New Zealand in response to an acute housing shortage and affordability problem. 

When Auckland Council wrote its zoning plan in 2016, the National government successfully pressured the local government to allow intensification across its non-historic suburbs by threatening to use their powers to overrule the council. And then in 2020, the Labour government forced the five largest cities to allow six-storey buildings within walking distance of city centers, commercial hubs, and existing and planned rapid transit stops. But the biggest change came in 2021, when both major parties joined together to force local governments to allow three-storey townhouses in almost every neighborhood in the five largest cities – encompassing at least 57 percent of New Zealand’s population. But this may have been a step too far. Implementation of this most ambitious policy proved fraught, with some local governments strongly resisting the top-down directive. In the face of an upcoming election and the shifting sands of political incentives, the cross-party consensus broke down.

New Zealand shifted from a liberal zoning regime to a more restrictive one in the 1970s as local communities were empowered to control development in their areas. Like in the US and the UK, local governments imposed restrictions to contain growth 'out', 'up', and protect historic character areas and green areas with development controls. Simultaneously local consultation requirements were made more elaborate and it was made easier to appeal planning decisions in the courts. This down-zoning led to the zoned capacity for new housing in Auckland being cut by half. 

Some of the zoning restrictions were idiosyncratic and unique to New Zealand,

There are ‘viewshafts’ that stop any developments that would block views of some of the 53 volcanoes in the region, including developments in the city center. One viewshaft over the central city, E-10, which makes sure Aucklanders can see Mt Eden from a particular point on a motorway, was estimated to block NZ$1.4 billion (US$800 million) worth of buildings. 

As has been seen with the cases of Houston and Washington DC, local governments will permit increased supply if there are financial incentives to do so like additional tax revenues. New Zealand has a structural problem with creating this incentive.

Unlike local governments in the US, Denmark, and Australia, which tax a fixed percentage of a property’s value, New Zealand councils tax via a rating system. Councils determine the total amount of revenue to collect each year, and then each homeowner pays in proportion to their share of total property value. In other countries, more housing automatically brings in more revenue; in New Zealand, new housing only slightly reduces the burden on existing properties.

New Zealand's breakthrough in upzoning came with the Auckland Unitary Plan (AUP), under which 75% of residentially zoned lands were upzoned, mostly meaning that it now allowed townhouses and even apartments. It was a big success with a large increase in dwelling consents and permits for new housing doubling within five years. 

Buoyed by the success of AUP, the New Zealand government sought to expand the upzoning strategy across the country.

The proposed National Policy Statement would require councils to allow at least six storeys of building in areas within walking distance of rapid transit stops, commercial hubs, and city centers in all major cities. It would also abolish minimum car parking requirements across New Zealand’s cities, which had required that new developments set aside large amounts of land for parking. Councils would need to justify character protections on a site-by-site basis, balanced against the need for more homes... The final policy was published in July 2020, less than a year later – uploaded to the internet with no fanfare, no announcement, no press release, to take effect the next month... Though the new upzoning policy avoided public scrutiny to begin with, it attracted plenty in late 2020 when the capital, Wellington – a city where historic-character areas protected 88 percent of land parcels in the inner-residential zone – began updating its zoning plan to reflect the new policy. Wellington was the first city to test the implementation of the new policy... In the end, character areas were reduced by 30 percent and upzoning was applied more widely than initially expected...

The Medium Density Residential Standards (MDRS)... policy (passed in December 2021) was even bolder than the first, directing councils to set a new default minimum residential zone in the five largest cities of three dwellings, up to three stories. Essentially, both major parties wanted to permit townhouses and walk-up apartments almost everywhere... This policy would go much further than the earlier National Policy Statement, covering nearly every suburb in the country, rather than just constrained areas in city centers and around rapid transit stops... opponents stressed that the upzoning policies would add pressure to existing infrastructure and raised doubts that the central government could address the problem. A wave of conservative and anti-development politicians were elected to councils around the country in October 2022 amid growing dissatisfaction with the prospect of densification.

The MDRS has struggled to get implemented in the face of opposition from local governments and the collapse of the bipartisan national level support among the two main political parties. The essay points to how the New Zealand government could have embraced the strategy adopted by Houston (more later) by allowing the most vocal pockets of opposition to upzoning to opt-out of broad upzoning instead of implementing the entire policy nationwide.

The Works in Progress has a story on Houston which has been a remarkable success in urban planning, especially in overcoming local resistance to easing zoning regulations.

Real house prices in Houston tumbled along with local incomes during the late 1980s, when the local oil industry took a beating from a global oil glut. But, remarkably, house prices and rents in Houston have remained low, even as the metropolitan area population increased from 2.4 to 6.7 million and the economy returned to booming. By 2021, they were still 23 percent lower in real terms than their 1980 peak. By contrast, inflation-adjusted prices in New York increased by 147 percent and in San Francisco by 216 percent, rendering those cities increasingly inaccessible for those on lower and indeed even higher incomes.

This is what Houston did with its zoning rules

Houston’s Code of Ordinances – its planning rulebook – sets limitations like minimum lot sizes (how big a plot of land must be used for each residential property), how far the building must be set back from the street, requirements for the number of parking spaces that must be included with the development, and much else. Historically, this set of regulations, which require a lot of land for each property, combined with the city’s permissive attitude to building outwards, led to Houston’s characteristic sprawl.

In 1998, a major change was made to the Code of Ordinances. The minimum lot size of a plot within the I-610 motorway which circles Houston’s inner suburbs was dropped from 5,000 square feet – about the size of a professional basketball court – to 1,400. This allowed landowners to divide (or ‘replat’) existing lots into much smaller parcels. The reform also changed the rules around property setback from the street, reducing the minimum required from 25 feet to as little as five feet. Whereas previously a landowner with a 5,000 square feet parcel would have generally been constrained to build one home, set back far from the street and with a lot of land surrounding it, now they could build three homes, for three times as many families...

Builders had started putting forward more frequent applications for replats into smaller lots in the more central suburbs. Initially the city’s planning authority had considered these on a case-by-case basis... So, rather than continue to process large numbers of replat applications one-by-one, the city redesigned the Code of Ordinances to enable these replats by right, removing the need to apply for each individual replat... The reforms have survived to this day – 2023 marks the twenty-fifth anniversary of their implementation – and, as further evidence of their popularity, were even extended into the suburbs outside of the I-610 motorway in 2013, meaning they cover the entirety of the city’s 671 square miles. 

It's outcomes have been impressive

The 1998 reforms, alongside Houston’s generally liberal approach to planning, helped to preserve housing affordability in the central neighbourhoods of the city, with the median townhome costing an estimated $313,000 in 2017, and the twenty-fifth percentile at just $156,000. Life is better for renters as well, with thousands of apartments available for less than $900 a month, unimaginable in many other cities within the US... Houstonians live in bigger, better housing than New Yorkers or Californians, and they do so much more affordably. Increasingly, they do so in pleasant, walkable environments... Indeed, what was once ‘renowned worldwide as the global citadel of carbon-based fuel extraction’ is now an ‘unexpected and almost entirely unheralded success story’ of urban renewal and better environmental planning... Houston, synonymous in many people’s minds with car-centric America, is now mid-ranked among American cities on the Foot Traffic Ahead walkability ranking, with its substantial improvement acknowledged in the latest report. In the 1990s it was the only major city in the United States without a rail system; it now has three light rail lines operating and two more planned.

Any local opposition to deregulation and easing of zoning laws was overcome by allowing homeowners associations (HOAs) covering small geographical areas to opt-out of city-wide zoning regulations or changes to those. This meant that nobody could hold hostage the easing of zoning regulations.

This opt-out system may explain why city-wide reforms were able to pass, and why they have had the staying power to last several decades relatively uncontroversially... Houston sets relatively limited rules about land use at the city level. Instead, many rules around land use are set within private ‘deed restrictions’. These are private agreements between landowners within blocks or small areas. When you buy a home in Houston, it may come with deed restrictions determining how you can use the land... If violated, deed restrictions in Houston can be enforced by the city at the request of neighbours, normally via their homeowners’ association, or HOA... 

Because these deed restrictions can specify if the lot has additional restrictions to the minimums set by the city-level code, Houstonian homeowners already had an institutional mechanism if they wanted to prevent new types of development near them, regardless of what happens in the citywide Code of Ordinances. If they didn’t want smaller lots to be permitted in their area, they could agree collectively via the HOA to add new deed restrictions. In 2001, new legislation was passed to make opting out even easier, by allowing local homeowners to petition the city to introduce a special minimum lot size (SMLS) even without going through the legal process of altering the deed restrictions... The petition needs to attract 51 percent support from local homeowners, or less than 51 percent support and no objections against... As a result, Houstonian homeowners who really don’t want new homes built near them don’t need to worry as much about the overall citywide rules. They have a much easier route nearer to home: they can agree with their neighbours in the HOA to alter the deed restrictions, or they can petition for a special minimum lot size... Houston’s system allows homeowners to opt out without bringing housing supply in other areas down with them.

And this has a self-regulating dynamic

What this means in a practical sense is that a Houstonian homeowner can opt to have more stringent restrictions on what gets built nearby, but it affects their own property in an obvious and direct way. If they want to set a minimum lot size of 5,000 square feet in their block, they can; but their own property will be worth less should they choose to sell it in the future. So they must decide: what’s worth more to me, the additional value to my property from the right to replace one home with two, or preventing my neighbours from having the same right? Different people come to different conclusions here. Houston allows them to, and it turns out many people don’t mind development next door as long as they have the same option. The benefits, and the costs, of additional regulation are therefore better aligned.

Another feature of Houston's zoning regulations is the sunset clause to private deed restrictions

Restrictions on land use in Houston typically have sunset clauses of 25–30 years. This is a smart workaround for the problem that many planning systems face, where stacks of rules, often conflicting or at least not designed to operate holistically, are added incrementally... Sunset provisions are used widely across legislative and regulatory fields, and they can be highly effective: states that use sunset programmes reduce their spending at the state level but increase the level of government services provided, suggesting that the sunsets help to weed out bad policies while preserving good ones. Similar benefits were seen in London’s Great Estates, where lease renewal dates were strategically timed to permit rebuilding. Much like how the Houston opt-out system prevents individuals from imposing their own preferences on neighbours who don’t share them, sunset clauses do this for future residents. In other places, a restriction preferred by a few residents can not only be imposed on much larger areas, it is also imposed on them indefinitely, long after the residents with the preference are gone.

In another article in Works in Progress, Judge Glock examines the reasons for the growth in regulations that stifle the expansion of supply. 

The best explanation for increasing regulation is that local governments and neighborhoods are less likely to see the gains from growth than they were in the past. Local governments used to get substantial fiscal benefits from a growing population, yet now they suffer fiscal costs. Local homeowners used to see the value of growth in terms of reduced taxes and increased land values, and now they see the increased taxes and worse services. States and local governments used to prepare infrastructure before growth happened, but now they let existing residents suffer crowding of roads and schools when new residents arrive.

The essay points to several examples of fiscal zoning (or sharing the benefits of new developments with local governments and local people) from across the world. It also points to a decline in fiscal zoning over time (restrictions on the local tax levy, reduction in sharing of revenues with the local government etc), contributing to rising opposition and NIMBYism in the US. He examines the reasons for increasing regulation,  

One reason is that in most communities the ‘winners’ from increased development are always small. There are only so many plots that can be built up at any time. The vast majority of neighbors only see the negative consequences of someone else developing their land and try to prevent those individuals from developing, even if they could later benefit from developing their own land. These neighbors can indeed increase the value of their own property by limiting development options on others property, but the end result is lower total property values in a community.

For much of the twentieth century, and in much of the United States today, the main way the developing plots compensated their neighbors was by providing extra fiscal revenue to the rest of the city. Although neighbors saw the costs of increased development, they also saw reduced taxes and increased services. Cities engaged in ‘fiscal zoning’ to attract the sort of development that brought extra funds and this was the impetus for increased local growth in general.

Glock's argument is to re-establish the fiscal benefits and make it obvious to residents that the costs of restricting new developments are much higher than its gains. This was the basis for the New Zealand reforms that we discussed earlier.

Interestingly, he points out that the vast majority of US cities and towns do not suffer from any supply shortages, and that regulations are a problem only in a few large cities - New York, San Francisco, Boston, Los Angeles, Seattle etc. He also points to research that shows that regulatory burdens increase the closer we get to the central parts of American cities. Most worryingly, while restrictive zoning and regulation were low or non-existent almost everywhere, they've been growing since the late 1990s. 

One reason for the restrictions in large cities is that the marginal benefit from an additional dollar of investment is very small compared to those in smaller or less developed cities. In the former, the infrastructure and service delivery are very good, whereas the latter requires significant additional investments. To this extent the existing households in larger cities have lesser incentive in raising resources. 

The researchers use the difference between the market price and the actual cost of construction, with the gap being a measure of regulatory burden (or a zoning tax). They estimate this zoning tax using a clever trick that calculates the difference between the value that an existing homeowner puts on having a bit more land (the intensive margin) and the value that a builder places on the same amount of land with the right to build on it (the extensive margin). The difference, a measure of zoning tax, would be higher in highly regulated cities. If it's higher, then the existing homeowner would be encouraged to sub-divide and sell out the land at the intensive margin to the builder. In cities with restrictive regulations, the builders would bid up the prices of the lands at the extensive margin till there were no unexploited profit opportunities left for builders. 

In this context, building on the paper by Nagpal and Gandhi, it would be useful for researchers to undertake studies like this and this to determine the extent of zoning tax in Indian cities. Which Indian cities (or states) have the highest zoning tax and which have low taxes, and whether and how do these taxes vary within these cities?

Weekend reading links

1. Good interview of Indonesia's President Joko Widodo. Widodo's second term comes to an end and given the term limits, he must step down despite an approval rating of close to 80%. With its 270 million youthful population, Indonesia should count as a future economic superpower, a top-five global economy. This is a good summary of Widodo's tenure

If there are two words that have defined Widodo’s presidency they are infrastructure and nickel. When he took office, demand for Indonesia’s bountiful commodities had slackened against a global downturn, its infrastructure was underfunded and enthusiasm from international investors for the world’s third-largest democracy was weak. The one-time city mayor set to work. Widodo points to the highways he has built, the airports that have sprung up in remote regions and the boom in new dams, seen as essential at a time when climate change has led to drought in large parts of the country. “Before, we had 240 dams, now we have 301,” he beams. 

But it is his second term, starting in 2019, which has seen the most ambitious economic policies, in particular the creation of a domestic electric vehicle and battery supply chain. By banning the export of nickel ore in 2020, Widodo forced companies such as China’s Tsingshan, South Korea’s LG and Brazil’s Vale to set up more local factories if they wanted access to Indonesia’s abundant reserves. These factories were not only to refine nickel but also to entice more companies to build more of their supply chains in Indonesia. Defying a ruling by the World Trade Organization that the ban was unfair, Widodo has stuck to his policy and it has paid off. Indonesia’s exports, boosted by soaring commodity prices, hit a high of $292bn in 2022 though they have moderated this year as China’s economy slows.

For all those who preach free-trade and oppose industrial policy, Widodo's success with bringing value capture to nickel processing into Indonesia should be one more reminder. Instead of promoting trade policies that favour large foreign multinationals, institutions like the World Bank should be helping countries like DRC and Ghana capture a greater share of value from their primary produce. 

2. Martin Wolf feels that the tightening cycle has come to an end and, more importantly, easing will start sooner than what central banks are suggesting. Core inflation in both US and EU have been falling sharply towards the target. 

In fact, if the target is a band of 2-4%, or say 3% (both of which ought to be the inflation target for a variety of reasons), then we are already in that band.

On similar lines, the graphic of the trajectory of interest rates over the last four decades is interesting.

Interest rates are only slightly above the historic trend. If expectations are for a return to a 2% interest rate trend, then I worry that's recreating the conditions for financialisation and other distortions. Instead a more sustainable and historically valid normal interest rate would be in the 3-4% range. We are actually closer to that range now than the markets imagine. But not if the markets expect the old norm of 2%. 

3. Peter Goodman in the Times describes the rise and fall of the world's greatest joint venture, the economic relationship between the US and China. While it brought spectacular returns, it also left its costs

Chinese imports effectively boosted the spending power of the average American household about 2 percent, or $1,500, a year from 2000 and 2007, according to one study. Chinese goods pressed down American prices 0.19 percent a year from 2004 to 2015, another study found...

Those hurt by Chinese imports were concentrated and conspicuous. Once-thriving American factory towns sank into joblessness and despair, swapping restaurants and hardware stores for food banks and pawnshops. From 1999 to 2011, a surge of low-priced Chinese imports eliminated nearly one million American manufacturing jobs and two million positions throughout the broader economy, according to a paper by the economists David H. Autor, David Dorn and Gordon H. Hanson.

4. Interesting facts about inequality in India

The OECD made an assessment of the number of generations required to move from the bottom 10 per cent of the income distribution to the mean income level. This was done by examining the rate of change of income between father and son. According to this assessment, the transition time from the lowest to the mean level of income in India is seven generations, a level comparable to that in China and above that in Europe and the US... In India, according to the World Inequality database, this has widened sharply between 1990 and 2018, the latest year for which they present an estimate. Over this period, the share of the top 10 per cent in pre-tax income has gone up from 34.4 per cent to 57.1 per cent, while the share of the bottom 50 per cent has fallen from 20.3 per cent to 13.1 per cent. Note also that the top 1 per cent account for nearly half of the increase in the share of the top 10 per cent... A recent report from Azim Premji University... points out: “In 2004, over 80 per cent of the sons of casual wage workers were themselves in casual employment. This was the case for both SC/ST workers and other castes. For non-SC/ST castes, this fell from 83 per cent to 53 per cent by 2018, and the incidence of better quality work, such as regular salaried jobs, increased. It fell for SC/ST castes as well, but to a lesser extent (86 per cent to 76 per cent)”.

From the Bihar caste survey

The sub-caste wise data on poverty indicates that 42.93 per cent of the families belong to Scheduled Castes (SC), 42.7 per cent to Scheduled Tribes (ST), 33.58 per cent to extremely backward castes (EBC), 33.16 per cent to other backward classes (OBC), and 25.09 per cent to the general category (GC) of upper castes. When it comes to government jobs, the general category castes, such as Bhumihars, Brahmins, and Kayastha, had the highest share with 3.19 per cent of their population in government jobs. The corresponding figures for the EBC, SC and ST are 0.98 per cent, 1.13 per cent and 1.37 per cent, respectively. A measure of the differences in access to education is provided by the substantial difference in the percentage of people who are graduates in different caste groups — 14.54 per cent in the general category, 9.14 per cent in the OBC category, 4.44 per cent in the EBC category, 3.12 per cent amongst Scheduled Castes, and 3.53 per cent amongst Scheduled Tribes.

5.  Infrastructure operators are earning a greater share of their revenues from non-core operations,

Core aeronautical revenue share of India’s largest airport in Delhi reduced to 24 per cent in FY23 from 46 per cent in FY18, the year it became a public company. Since then, it has consistently earned more from non-aeronautical revenue streams including verticals like duty free, space rentals, and cargo, and commercial property development than from core revenue streams including landing charges, user development fee (UDF), and baggage x-ray charges. This is primarily due to low aeronautical charges levied upon airline operators for landing and parking aircraft... Indira Gandhi International Airport’s operator Delhi International Airport Limited (DIAL), in which the GMR Group owns a majority stake, generated an operating revenue of Rs 3,990 crore in FY23, of which Rs 938 crore was from aeronautical services and charges, Rs 2,477 crore from non-aeronautical services and charges, and Rs 575 crore from licence fees in connection with certain commercial property development activities at the airport. DIAL’s aeronautical revenue share dropped to 24 per cent in FY23 from 30 per cent in FY19 as the share of revenue generated through commercial property development increased to 14 per cent from 6 per cent between the same years. DIAL’s non-aeronautical revenue share has remained relatively stable, averaging around 58 per cent in the last five years, and includes revenue streams like duty free, land and space rentals, advertisements, cargo, F&B, and retail... 

In a similar trend, Mumbai’s Chhatrapati Shivaji Maharaj International Airport, operated by Mumbai International Airport Limited (MIAL), in which Adani Enterprises now holds a majority stake, also saw a drop in its aeronautical revenue share to 38 per cent in FY23 from 46 per cent in FY19 as its non-aeronautical revenue share jumped to 62 per cent from 54 per cent. In FY23, MIAL generated a total operating revenue of Rs 3,233 crore. As per Fitch Ratings, MIAL also uses 30 per cent of its non-aeronautical revenue for cross-subsidisation. For both Delhi and Mumbai airports, user development fee (UDF), which is a charge levied on passengers and is included in their ticket price, forms a large chunk of aeronautical revenue. The approved rate of UDF for DIAL by the Airports Authority of India is Rs 200 for domestic passengers and Rs 1,300 for international passengers, whereas for MIAL it is Rs 100 for domestic passengers and Rs 600 for international passengers. In FY23, Delhi airport saw average monthly traffic of 5.44 million passengers whereas Mumbai airport saw 3.66 million passengers... 

Similarly, Delhi Metro Rail Corporation (DMRC) has become increasingly reliant on external project income, which accounted for 48 per cent of operating revenue in FY22 compared to 35 per cent in FY19, while core revenue generated from passenger traffic dropped to 39 per cent in FY22 from 55 per cent in FY19... DMRC’s traffic revenue share fell to 39 per cent in FY22 from 55 per cent in FY19, while the share of external project income in total operating revenue jumped to 48 per cent from 35 per cent in the same time period... DMRC earned an average of Rs 45 crore every year between FY19 and FY22 by rendering consultancy and training services to various domestic metro projects in cities like Noida, Mumbai, and Jaipur, and to metro projects in other countries, like in Dhaka. In FY22, DMRC also raked in nearly Rs 500 crore through rental and lease earnings, which was slightly lesser than Rs 559 crore in FY19... 

IRCTC, incorporated in 1999, generated higher net profit from e-ticketing services than from catering and Rail Neer sales in FY23. Even though IRCTC’s core revenue from catering and the sale of its flagship bottled water product, Rail Neer, accounted for 50 per cent of total operating revenue in FY23, compared to e-ticketing revenue share of 34 per cent, it made Rs 1,021 crore in net profit from e-ticketing services compared to Rs 204 crore from catering and Rail Neer sales. IRCTC has the exclusive mandate for Indian Railways’ e-ticketing services, which began in 2002... In FY23, IRCTC’s e-ticketing revenue stood at Rs 1,198 crore and accounted for 80 per cent of total net profit, eclipsing IRCTC’s net profit from core revenue streams including catering, Rail Neer, and tourism services.

6. Times reports of a startup, Heirloom Carbon Technology, trying a new technology to remove carbon directly by vacuuming greenhouse gases from the atmosphere. This startup thereby takes the idea of removing carbon dioxide from air from science fiction to reality. 

Heirloom will take the carbon dioxide it pulls from the air and have the gas sealed permanently in concrete, where it can’t heat the planet. To earn revenue, the company is selling carbon removal credits to companies paying a premium to offset their own emissions. Microsoft has already signed a deal with Heirloom to remove 315,000 tons of carbon dioxide from the atmosphere. The company’s first facility in Tracy, Calif., which opens Thursday, is fairly small. The plant can absorb a maximum of 1,000 tons of carbon dioxide per year, equal to the exhaust from about 200 cars...

Heirloom’s technology hinges on a simple bit of chemistry: Limestone, one of the most abundant rocks on the planet, forms when calcium oxide binds with carbon dioxide. In nature, that process takes years. Heirloom speeds it up. At the California plant, workers heat limestone to 1,650 degrees Fahrenheit in a kiln powered by renewable electricity. Carbon dioxide is released from the limestone and pumped into a storage tank. The leftover calcium oxide, which looks like flour, is then doused with water and spread onto large trays, which are carried by robots onto tower-high racks and exposed to open air. Over three days, the white powder absorbs carbon dioxide and turns into limestone again. Then it’s back to the kiln and the cycle repeats...

The carbon dioxide still needs to be dealt with. In California, Heirloom works with CarbonCure, a company that mixes the gas into concrete, where it mineralizes and can no longer escape into the air. In future projects, Heirloom also plans to pump carbon dioxide into underground storage wells, burying it. Heirloom won’t disclose its exact costs, but experts estimate that direct air capture currently costs around $600 to $1,000 per ton of carbon dioxide, making it by far the most expensive way to curb emissions, even after new federal tax credits worth up to $180 per ton. Heirloom has set a long-term target of $100 per ton and aims to get there, in part, through economies of scale and mass-produced components. For its next plant, planned in Louisiana, Heirloom will use a more efficient kiln and a denser layout to save on land costs.

7.  The dramatic ouster of Sam Altman as CEO of OpenAI by the company Board is reflective of the rift in the AI community between those who see AI as the biggest business opportunity and those who are concerned at the existential, social, political and economic challenges posed by it. Or as FT put it nicely,

At the core of the recent turmoil is the tension between the commercial drive of any company to make money and concerns about the collateral damage that technology can cause.

The swift return and the support from Big Tech bigwigs and influential opinion makers like Larry Summers is equally emblematic of the dominance of moneyed interests over all ethical and other considerations. The episode has further empowered Altman as the messaiah of AI, and it's hard to imagine the new Board will exercise any oversight in terms social and ethical responsibilities about the direction of OpenAI's business. It's safe to assume that the first round has been decisively won by those who see AI in terms of the business opportunities it presents. 

Now that the company leading the AI charge and its leadership would be even more emboldened in pursuit of its technological and business ambitions and all restraints have been eliminated, it's more important that the government step in to regulate AI. A few individuals and one company cannot be allowed to have so much power in determining the direction of the leading technology of our times.  

In fact, the episode also ends the hypocrisy behind OpenAI's founding principle in 2015 of prioritising safety and humanity over profit. Under Sam Altman, whose roots are unambiguously in the world of techno-evangelism and profits, the presence of a non-profit Board overseeing a relentlessly profit-seeking and technology frontier expansion focused company was becoming untenable. 

Open AI now joins the list of companies with excessively powerful CEOs and very weak boards. We know what happens to such companies. But unlike others here there are existential consequences for humanity, and the race to the frontier of artificial intelligence looks like an unconstrained pursuit.  

8. Evidence of Covid 19 related learning loss in the US

The school closures that took 50 million children out of classrooms at the start of the pandemic may prove to be the most damaging disruption in the history of American education. It also set student progress in math and reading back by two decades and widened the achievement gap that separates poor and wealthy children... Economists are predicting that this generation, with such a significant educational gap, will experience diminished lifetime earnings and become a significant drag on the economy.

9. Couple of good suggestions to address the problem of stubble burning by farmers in North India that has been found to be a major contributor to the toxic smog in Delhi. The first is to replace free farm power with some form of DBT. The second is to create a market for paddy straw/stubble to incentivise farmers to harvest, collect and dispose it.

Baling machines (balers) for paddy straw are already in use in Punjab and Haryana, which has made it feasible to put paddy and other crop straws in the value chain. On average, an acre of land generates about 2.5 tonnes of paddy straw. The cost of baling this stubble is anywhere between Rs 1,000 to Rs 1,100 per acre (including the cost of cutting, raking, packaging and storing bales on land). From our estimates, it appears that a price of Rs 1,000-1,200 per tonne of bale of parali, covers the costs and leaves a small margin for the farmer. Punjab generates about 20 million metric tonnes (MMTs) and Haryana has about half of this. About 85 per cent of it is burnt in the field. Thus, the total cost of procuring the entire parali burnt in the field in Punjab comes to Rs 2,000 crore and, in Haryana, about Rs 1,000 crore. A small market for paddy straw sold in compact bales has already emerged in both the states for production of biofuel such as BioCNG and ethanol and as direct fuel in brick kilns, furnaces, and thermal plants. Some enterprising farmers have sold parali at Rs 180 per quintal this season — this can be treated as an indicative market value. This market is picking up but at its current pace, it will take many years to match the supply. 

Therefore, ways and means need to be put in place for adequate availability of balers and incentives introduced for the use of stubble as a biofuel. Among various options, the use of straw for the production of compressed biogas through methods of anaerobic digestion is best from economic and environmental perspectives. It also produces bio-slurry, which can go back into the soil to replenish soil fertility. It is a matter of supporting the supply chains of paddy straw initially for four to five years. This will set the trend for converting agri waste into wealth and promoting a circular economy in agriculture in the entire country.

10. Argentina cannot but not help inviting economic crisis very close frequency. The victory of radical libertarian economist Javier Milei in the country's Presidential election run-off is a recipe for instability. Milei defeated the establishment Peronist candidate and Economy Minister Sergio Massa wining more votes in the run-off than any candidate since 1983.

Milei’s campaign centred on a pledge to take a “chainsaw” to the state — slashing spending by up to 15 per cent of gross domestic product — and to dollarise the economy to stamp out inflation. Argentina’s annual price rises hit 142.7 per cent in October... Milei, a self-described “anarcho-capitalist”, stirred controversy throughout the campaign, expressing support for ideas such as legalising the sale of human organs and eliminating all gun laws. He also referred to China, Argentina’s largest trading partner, as “murderous”, the Argentine Pope Francis as “a filthy leftist” and climate change as “a socialist hoax”...

The win for Milei, a former television commentator who became famous for rants against economic mismanagement and corruption among Argentina’s governing elite, is a rebuke against Massa’s Peronist movement, which has dominated politics since the country returned to democracy in 1983. Over the past two decades, left-leaning Peronist governments have doubled the size of the public sector and introduced expensive subsidies and tight regulation across the economy. The Peronist model has faced unprecedented pressure this year amid spiralling inflation. Massa has resorted to money-printing to finance spending and tightened strict trade and exchange restrictions to protect scarce foreign currency reserves...

Most economists in Argentina say Milei’s flagship plan to replace the peso with the US dollar is unworkable in the short term given that Argentina has almost no dollars in its central bank and no access to international credit. The official exchange rate is fixed at just over 350 pesos to the dollar, but the black-market rate is as high as 900 pesos. The gap, which has widened dramatically as the parallel exchange rate has plunged in recent months, has caused widespread distortion of prices.

This is a stunning statistic about Argentina's dollar use

About 10 percent of all U.S. currency in circulation is in Argentina, according to some estimates, or roughly $200 billion, more than in any other country outside the United States. That’s an average of about $4,400 in cash for every Argentine, compared to $3,100 for every American... In April 2020, at the start of the pandemic, $1 bought 80 pesos at the “blue dollar” rate. A year ago, $1 bought 300 pesos. On Tuesday, as markets in Argentina opened for the first time since Mr. Milei’s victory, the peso’s value fell to a record low. That day, $1 bought 1,075 pesos.

The challenge Milei faces is to dollarise without owning dollars, as the government's dollar reserves are miniscule.  

11. India is witnessing an explosive growth in the Global Capability Centres (GCC) of multinational corporations and financial institutions, back-offices managed by the company itself that develop in-house technology, including cyber security systems, artificial intelligence, accounting and HR work, R&D in general. 

Global capability centres have proliferated and expanded at a rate of 11 per cent a year since 2015 to become a $46bn industry employing 1.7mn people in India, according to Nasscom. Real estate group Colliers estimates the number of GCCs in India will double from 1,026 in 2015 to 2,000 by 2026.

This off-shoring of the company's work has unleashed a competition for talent with the Indian IT outsourcing majors. That's a great thing to happen and should be one Moree reason why these IT firms should buckle up and move up the value chain.

12. Novo Nordisk's magic drugs Wegovy and Ozempic are on a roll. From diabetes risk reduction to weight-reduction, they have now been tested to reduce the chance of death from heart attack or stroke by 20%.  

Not since the rise of cholesterol-reducing statins, or perhaps even pain medications like Advil, has a group of pharmaceuticals so captured the public imagination. Wegovy, and its better known cousin Ozempic, are “semaglutides,” a class of drugs that slow digestion and mimic the effects of natural appetite-reducing hormones. First commercialised by Danish insulin maker NovoNordisk, they are now being developed and rolled out by many major pharmaceutical companies. Not only do they lead to an average 15-20 per cent weight loss in obese patients, they also appear to protect the heart, liver and kidneys, organs which are often put under strain by excess weight. Prescriptions for these drugs are up a whopping 300 per cent in the US since 2020, despite the fact that they can cost between $300 and $1,300 per month. Bank of America expects 48mn Americans (about one-seventh of the population) to be on the meds by 2030.

These drugs are disrupting several industries, including Krispy Kreme!

Let’s start with the pharmaceutical firms themselves. If you don’t have an Ozempic knock-off in the development pipeline, your share price may take a hit. Novo Nordisk now has a market capitalisation that is higher than the entire gross domestic product of Denmark, and Eli Lilly’s share price is up 40 per cent since it rolled out its own weight-loss copycat Mounjaro. But both Pfizer and Moderna — neither of which have a successful semaglutide on the market — have seen their share prices plummet in recent months... In early October, when Novo Nordisk announced that Ozempic was so effective against kidney disease that it was stopping a trial early, shares in some dialysis providers tanked. Now, healthcare analysts say that the $250bn cardiovascular disease market could be reduced by 10 per cent by 2050, and hundreds of billions-worth of additional business in treatments for diabetes, kidney and liver disease and other weight related illnesses could be disrupted. The Ozempic effect doesn’t stop there. Analysts have downgraded doughnut maker Krispy Kreme recently amid worries that Americans on semaglutides just won’t reach for as many sweet treats as they have in the past. Last month, Walmart chief executive John Furner said that customers on obesity drugs weren’t buying as many groceries, which led to a brief sell-off in consumer staple stocks such as Mondelez and PepsiCo.

13. With the RBI taking note and raising the risk weights on consumer loans, the issue of rising consumer loans in banking portfolios has taken centre stage. Unsecured loans have been growing at 23% in the last two years, compared to lending to corporations and SMEs at 12-14%. Tamal Bandopadhyay writes

According to the RBI’s latest Financial Stability Report, released in June, advances for unsecured retail loans rose to 25.2 per cent in March 2023 from 22.9 per cent in March 2021, while secured loans saw a decline to 74.8 per cent from 77.1 per cent during the same period. Overall, retail loans grew at a compound annual growth rate, or CAGR, of 24.8 per cent from March 2021 to March 2023 in contrast to 13.8 per cent CAGR for overall bank credit... Delinquencies, measured in terms of the inability of borrowers to repay loans between 31 and 180 days, for loans under Rs 50,000 rose to 8.1 per cent in June 2023... the total value of short-term personal loans (STPL) below Rs 10,000 grew 37 per cent in the financial year ending March 31, 2023, while STPL of Rs 10,000-50,000 rose 48 per cent. The credit bureau considers loans up to Rs 50,000 as STPL.  Some 8.6 million such loans were disbursed in FY23, registering a 50 per cent rise over FY22. About 80 per cent of all personal loans disbursed in FY23 were STPL and 60 per cent of such loans had an ultra-small ticket size — below Rs 10,000. The overall personal loan portfolio was to the tune of Rs 11.16 trillion as of June 2023 — more than double of the level seen in March 2020 (pre-Covid).

Small cities are contributing more to these loans. About 38 per cent of STPL up to Rs 10,000 in the last 12 months were outside India’s top 100 cities. In contrast, 29 per cent originated in the top eight cities. To support the fact that small cities are playing a bigger role in this segment, the credit bureau says 35 per cent of STPL between Rs 10,000 and Rs 50,000 sanctioned between July 2022 and June 2023 were from beyond the top 100 cities, while the top eight cities accounted for 31 per cent. NBFCs dominate the origination and the portfolio of such loans. The share of private banks by origination volume has risen from the pre-Covid level, but relative to March 2022, it has dipped... Since January 2022, small-ticket personal loans of less than Rs 50,000, while representing a very small share of total retail balances, have roughly accounted for one-fourth of total volumes. In the June quarter of FY24, 51 per cent of consumers who took small-ticket personal loans already had more than four credit products at the time of accessing yet another new loan, compared with just 17 per cent in the June quarter of FY20.  

As Tamal writes, there's clear stress at the bottom of the pyramid.

14. Graphics on fossil fuel consumption trends over the last decade.

15. Finally TN Ninan points to the pricing problem that Indian Railways faces. The organisation is struggling to increase the ticket prices on its passenger trains, which are low compared to even bus fares. 

UP Roadways will take you 550 kilometres from Delhi to Lucknow for Rs 822, or Rs 1.49 per km. The Railways will get you there for barely half the price, Rs 432. Indeed, you could get there in air-conditioned comfort for Rs 755 in a three-tier sleeper coach — less than what the “Volvo” bus service charges: Rs 1,000 upwards for “semi-sleeper” travel. Across the railway system, the passenger fare per kilometre ranges from as little as 21 paise for “ordinary” second-class travel to Rs 1.75 for a seat in an air-conditioned chair car, and more for greater comfort or speed (eg: Rs 2.58 by the Shatabdi). Across all classes of travel, the Railways earns Rs 2 per passenger-km... With expenditure more or less matching annual revenue of Rs 2.65 trillion, most of the massive investment under way has necessarily been funded by borrowings and (because debt-servicing costs have grown) increasingly through support from the Budget. The current surge in investment is outsize: Over the past decade it has crossed Rs 13 trillion, with Rs 2.6 trillion this year alone being almost equal to projected revenue!

The low fares in turn has generated a set of perverse incentives. 

The demand for rail travel far exceeds supply, but the Railways has no financial incentive to increase passenger transport. The attention given to increasing haulage capacity has focused on new freight corridors, less so on boosting passenger transport capacity. The shocking result is that, while freight traffic has grown 40 per cent in a decade, passenger traffic has been static. In fact, it is marginally lower this year than a decade back... To make good the loss on its passenger service, the Railways pushes up freight rates for goods transport (the old cross-subsidy game loved by all populists). This is said to explain why the Railways has lost out on goods traffic to road, accounting now for only a quarter of the total, with almost all of it being bulk items like coal and iron ore.

Real estate related public revenues in China

Land has been an important factor in China's spectacular economic growth. Its ownership by the government has been critical in helping local governments realise revenues to finance the massive infrastructure investments that turbocharged growth for decades. The government generates revenue from land by way of auctioning or tendering land to private people and firms, taxation on various land related activities (transfer deeds, development fees, property taxes etc), and taxes from real estate firms. 

Martin Sandbu points to this excellent paper by Sheng Zhongming that highlights the importance of land to the Chinese economy. This is a great primer from a local source about the importance of land and real estate on the country's economy and public finances. 

The paper discusses the large effect of a fall in real estate activity on public finances.

After recent adjustments, the annual sales of commercial properties have declined to around 13 trillion yuan, which is considered a sustainable level. The article estimates that if real estate sales remain stable at the 2022 level, there will be a decrease in tax revenue by 0.8 trillion yuan per year compared to the pre-adjustment period. Additionally, land transfer income will decrease by 2.8 trillion yuan annually, and government borrowing, including special bonds and debts of urban investment platforms, will decrease by 3.6 trillion yuan annually. In total, this results in a decrease of 7.2 trillion yuan in fiscal revenue and government financing...

Looking at different regions, the absolute decline in fiscal revenue and financing is larger in eastern and central provinces. Guangdong, Zhejiang, Jiangsu, and Shandong will experience a combined decrease of 700 to 800 billion yuan in local fiscal revenue and government financing. The relative decline in fiscal revenue and financing is higher in central, western, and northeastern provinces. If we put transfer payments aside, some provinces will have to bear a fiscal revenue and financing decline of 30% or more.

Sandbu writes that this reduction in public sector revenues and available government financing by Rmb3.6 trillion each corresponds to 3-4% of GDP each. 

Most of this will hit local government budgets, which in 2021 made up about two-thirds of overall fiscal revenue of Rmb30tn. In other words, we could be looking at a permanent revenue loss of 15 per cent, and as much again in curtailed financing, for that government level.

However the paper points to the functioning of land markets and different kinds of revenues from land-related activities. I'll do a graphical summary of the paper. This is a good summary of the importance of real estate activities.

The importance of real estate in fiscal revenue and government financing is mainly manifested through three channels: specific tax types that contribute to real estate industry tax revenue, high-priced real estate land supporting land transfer income, and leveraging land value to drive government financing.

A very high 36.4% of the total budgetary fiscal revenue in the country came from the real estate sector (commercial land development), including 16.9% of all tax revenues and 92.1% of all land transfer income. While only 9.3% of central fiscal revenue and 9.9% of central tax revenue came from real estate, it made up 49.1% of local fiscal revenue. Further on the financing side, issuance of local special bonds and urban investment bonds is highly dependent on land value. 

Budget revenue consists of the general public budget (includes tax and non-tax revenues) and the government fund budget (which includes land use rights transfer income and more than 20 other government fund revenues). 

Total direct tax revenue related to real estate touched 2.9 trillion yuan, of which 2 trillion yuan belongs to local governments. 

Land value tax and title transfer (deed registration) tax make up 41.2% of all real estate related tax revenues, while property tax makes up just 3.6%! The first two go completely to local governments. 

The vast majority, nearly 90%, of the land value tax is paid by real estate firms.

In recent years while the proportion of land allotted for real estate development (commercial and residential) has declined, its unit price has risen sharply to ensure that the proportion of land transfer revenues from real estate has remained stable at a very high 90% of all land transfer revenues.

Now let's look at how land is leveraged to raise debt. The paper has a good description of the origins of the land revenue related infrastructure projects financing model

In 1998, the Wuhu Construction Investment Corporation bundled six infrastructure projects and signed a 10-year, 1.08 billion yuan loan contract with the China Development Bank. The loan was pledged against land transfer revenue, and the Wuhu Municipal Finance Department provided repayment commitments to enhance creditworthiness. This infrastructure financing model, which is based on land value, relies on government credit as explicit or implicit guarantees and does not create explicit government debt, became known as the "Wuhu Model" and served as a prototype for subsequent urban investment platform operations. Furthermore, since the promulgation of the new Budget Law in 2014, local governments have been granted autonomous borrowing authority, with special bonds used for projects with stable returns. However, reviewing the fundraising materials for various special bonds reveals that the expected land revenue held by project entities is a significant source of repayment for many special bonds.

The paper points to the land-to-government financing multiplier ((new special bonds + new interest bearing urban investment bonds)/land transfer revenue) of around 1.3 - at the national level, for every 100 yuan of land transfer revenue, nearly 130 yuan of broad government financing is unlocked! 

Local government fiscal revenues which consist of three parts (general public budget revenue, government fund budget revenue, and central transfer payment revenue) present a fascinating picture. There is a near perfect inverse correlation between the local government's real estate related revenue and central transfer payments. 

The paper categorises the provinces into four groups - those that have experienced land market decline far greater than others (Type I), those where the risk of both government revenue and financing being affected is high (Type II), those where government financing being effected is low but government revenue impact is relatively high (Type III), and those where both the risks of government financing and revenue impact is low (Type IV). 

Finally, the paper points to the inflection in the real estate market that saw a sharp decline in real estate construction activity and sales.

Based on this, the paper points to the base case (real estate activity stabilises at the 2022 level), optimistic, and pessimistic scenarios, which point to a combined drop in fiscal revenue and government financing by 7.2, 5.8 and 8.7 trillion yuan respectively. These are huge drops and their impact on economic activity has to be significant.