Mass transit and land use planning drives economic clusters

Large urban agglomerations are engines of national economic growth. The fuels that power these engines are the transportation infrastructure and mass transit, supported with enabling land use and development control regulations, value capture and revenue sharing, and institutional architecture. This is the topic of the post.

Such urban agglomerations are characterised by clusters of dense, specialised nodes of employment connected to residential nodes by commuter networks. This is the classic model of metropolitan downtowns connected with densified suburban satellite towns that coordinates agglomeration economies at the centre for high productivity job creation.

Accordingly, the central parts of major metropolitan cities of the world are predominantly workplaces, with people commuting to work from suburban areas on commuter railway systems. Mumbai, London, and New York are great examples. This is in keeping with the economic logic of ensuring the most productive use of the highest value lands in the city.

It also unlocks affordable housing supply by locating them predominantly in those areas with lower land prices, thereby relaxing the biggest constraint to affordable housing.

The ingredient that binds the nodes together and creates a dynamic regional cluster is a well-connected mass transit network. In cities globally, commuter and metro railway systems have played this role.

High-volume mass transit systems connect the nodes and bring people closer to jobs. They reshape the geography of what an hour reaches. This trip-shortening effect — making longer distance commutes faster than shorter car trips — is especially powerful in India, where the spatial mismatch between jobs in metropolitan cores and job-seekers in the hinterland has been found to be unusually large. They shrink the distances, thereby increasing the pool of jobs available for people in the hinterland, and expanding the talent accessible for employers in the core.

But evidence also informs us that merely laying the mass transit network is not sufficient. This study by Rana Hasan et albuilt a spatial model of Bengaluru and run the metro against alternative land-use policies and came up with the following quantified impacts. 

Holding the metro investment identical, the difference in net benefits between the best and worst land-use plans is roughly 32 percentage points of city GDP. In other words, the land-use plan determines whether the transit investment creates value or destroys it.

The requirement is to locate sufficient density of homes and workplaces within conveniently accessible distance from the stations, either through walk or feeder services. This is transit oriented development (TOD).

TOD is about commuting using mass transit by walking to/from station (to/from home and workplace) or feeder services. A big collateral benefit from TOD is that it avoids the use of cars, thereby decreasing traffic congestion, reducing productivity losses from long commutes, and lowering air pollution.

As I blogged here, TOD also increases the commercial viability of the massive investments made in mass transit systems (~ Rs 8 lakh Cr and counting, with 20-30% ridership realised, and massive debt repayment fiscal burden looming). It induces traffic by creating new trips (by locating more people within the transit zone influence area), shortening trip distances (thereby expanding the effective commute catchment), and mode shifts (from cars to mass transit).

Effective implementation of TOD requires addressing three constraints. One, relaxing development control regulations to make it attractive to build density in the vicinity of the stations, and creating the complementary pedestrian infrastructure. It also requires appropriate mixed land-use planning that allows the market to respond in the most efficient manner to order commute patterns and shape spatial growth. 

Two, funding of the mass transit capex, especially by enabling with the right kind of value capture instruments, with the right sequencing, and also the appropriate revenue sharing between the rail company and the local government/UDA. Try to capture value too quickly, before value gets created, and the cluster remains still born.

Three, an institutional architecture that aligns the incentives between those constructing the transit system (and repaying the debt mobilised) and those responsible for the land use and other decisions that determine the commercial viability of the system. This could, for example, entail the radical change of making the urban development authorities the owner of metro railway systems in metropolitan areas.

Observations on India’s external exposure in the age of capital wars

There are two important trends that are likely to impact cross-border capital flows for the foreseeable future. The first is the already high and increasing public debt burdens and persistently high budget deficits across the developed world. The second is the growing efforts to reshore manufacturing and reduce the dependency on China. The US is at the forefront of both these trends. 

Both are having large and growing spillovers across the global financial markets, especially adversely impacting developing countries. The first trend is raising sovereign bond yields, forcing interest rate hikes, and weakening currencies. The second trend is triggering a rising competition to attract manufacturing investment and retain domestic savings. 

I blogged here arguing that these trends will intensify the competition for capital, and most likely trigger capital wars. I had also discussed the possible implications for India. 

These trends come at a time when India is continuing the opening up of its financial markets and actively courting foreign investment. What do these trends mean for the RBI’s monetary policy, the rupee, India’s foreign direct investment, foreign portfolio investment, external commercial borrowings, and bilateral and multilateral aid flows? 

Finance 101 and historical experience tells us that rising US Treasury yields trigger capital flow reversals, puts pressure on the rupee, worsens the trade deficit, and reduces RBI’s monetary policy autonomy. And there is a self-reinforcing spiral associated with this cycle. On top come the squeeze on FDI inflows arising from the efforts to retain domestic savings and reshore manufacturing. 

Globally, the rising stock and flows of public debt portend a long period of higher interest rates, also because of the inflationary pressures that are already on play and will remain so for the foreseeable future. 

In this backdrop, this post will use data analysed with Claude to examine the consequences of India’s external exposure. 

Arguably the biggest deterrent to foreign capital inflows into India today is the weakening trend of the rupee. A weaker rupee raises the hurdle rate that any investor, FDI or FPI, would demand for investing in India. The rupee depreciation, therefore, functions as a direct tax on foreign investors' dollar-denominated returns. Every 1% of depreciation eats roughly 1% of the foreign investor's return. 

As an illustration, the rupee has weakened from ₹74/$ in 2021 to ₹96/$ in 2026, a cumulative decline of 30% over five years. This means that any foreign investors who remained invested in India should have earned an additional 30% returns to offset the depreciating currency. 

So what drives rupee depreciation? 

The obvious candidates are inflation differential, US Federal Funds rate, current account deficit, and oil prices. While a contemporaneous model OLS regression of annual rupee depreciation on these four structural drivers barely fits, a one year lagged model fits well (R-sq = 0.62, F-stat p = 0.001). The rupee depreciation correlates primarily with oil prices (the import-bill channel), the inflation differential (mediated by the RBI’s response), and the current-account balance, but with a one-year lag. 

Going by this, given the prevailing global market conditions and with no signs of an end to the Iran war, the rupee is primed for further declines in the year ahead.

The prospects of further rupee weakening is reflected in the USD/INR forward premium, which compressed dramatically from ~5% in early 2021 to barely 1% by mid-2024 as Fed hikes narrowed the India-US rate gap, but has since been rising and is about 3.5-4.1% in October 2026. For ECB borrowers, the forward premium is the hedging cost.

What does all this mean for RBI’s monetary policy choices? 

The Uncovered Interest Parity (UIP), which is an unobserved theoretical construct, states that the difference in interest rates between two countries should equal the expected change in exchange rates between their currencies, so that there is no arbitrage opportunity. The reality though is complicated by risk premiums and transaction frictions. 

The higher-for-longer global interest rates narrow the monetary policy space to cut or even maintain neutral stance in India without triggering outflows and rupee depreciation. In an open-economy framework where India’s capital account is substantially open on the debt side, the UIP floor on the repo rate rises mechanically. And, as the graphic below shows, it has been rising since 2021. With the Fed at 4.25% and the BoJ no longer anchored at zero, the UIP floor on the repo rate is around 6.3%.

This has created a situation where the UIP floor today is above the RBI’s repo rate. This puts pressure on the rupee to depreciate, something backstopped only the RBI’s large reserves. In fact, the RBI’s 5.5% repo rate works today only because of the forex-reserve buffer. 

However, now with US looking at the beginning of a rate hike cycle, the RBI’s room for monetary policy autonomy is considerably squeezed. It has little choice but to respond with raising rates. 

The capital wars channel compounds the problem. A big risk is the US net Treasury issuance estimate of $2tn/yr through 2028. It will be a major sink for global savings, and will raise the hurdle rate for all EM sovereign and corporate issuance.

An associated signature of this squeeze is the narrowing of the spread between Indian and US 10 year sovereign bonds. It has more than halved since the beginning of 2021. Coupled with the currency hedging cost, the spread disappears, thereby eliminating any incentive among those pursuing carry trade opportunities. 

What does this imply for India’s external debt exposures? 

In the aggregate, India’s external debt looks moderate at 20.8% of GDP. But there are two dimensions that need monitoring. The first is that the historical access to cheap, long-tenor, concessional fixed rate debt, primarily from bilateral and multilateral institutions, has now given way to floating rate loans benchmarked to SOFR/TONA/TORF (the transition from LIBOR happened by 2023). Of the $174.3 bn sovereign debt exposure (as of June 2026), 58% are now floating rate loans. More on this latter.

The second is the composition of external debt is now shifting away from the government to the private sector. This composition deserves closer monitoring given the rising share of External Commercial Borrowings (ECBs). 

At 32.8% of India’s external debt, ECBs are now bigger than the combined value of multilateral and bilateral loans and NRI deposits. The downward pressure on the rupee is a major risk for ECBs. This risk is amplified by the rising rates globally since most ECBs are floating-rate (SOFR-linked).

Of the $255bn outstanding ECBs, only 66% are hedged, and the remaining are direct, uninsured pass-through. The 34% unhedged slice ($87bn) is where the first stress will surface. It will fall most likely at NBFCs, airlines and renewable-energy developers that borrowed at low SOFR in 2021–22 and now must roll.

The impact of rupee depreciation on the $87 bn of unhedged ECBs is significant. 

These risks are also serious also because of the shorter repayment schedules of the ECBs. About $120 bn are due for repayment till March 2028. 

Another cause for concern may be the nature of ECB borrowers. An Indian corporate borrowing via ECB faces three cost layers - the base rate, currency depreciation, and the cost of hedging. The chart below shows that at least since 2023, ECBs should have been losing their attractiveness, especially for the higher rated borrowers. Nevertheless the ECBs have continued to rise, though issuance has fallen 33% in FY26. This raises a concern that some proportion of the ECB borrowers may be coming from the riskier market segments. 

The rising ECB exposure necessitates greater RBI oversight and microprudential regulation. For example, it should mandate higher ECB hedging ratios for sectors where revenues are in rupees (airlines, renewables, NBFCs). It should keep a close watch on higher cost commercial borrowings by by systemically important corporates.

The impact of rising interest rates globally is also showing up on India’s bilateral and multilateral borrowing windows. The notion that multilateral and bilateral lending is cheap, long-tenor and insulated from financial-market cycles is no longer true. ADB, World Bank and JICA have all transitioned from LIBOR to the new overnight benchmarks — USD SOFR, Yen TONA and the newer TORF. Each is 300–350 bps above its 2020 floor.

For four decades, yen funding has been India’s cheapest formal external finance. JICA has been the backbone of financing for Delhi Metro, Mumbai Trans Harbour Link, Delhi Mumbai Industrial Corridor (DMIC), Western Dedicated Freight Corridor (DFC) and the Mumbai-Ahmedabad bullet train. All of them were conceived when the all-in yen cost was below 1%. But BoJ’s rate hikes, at 1.25% in Sep 2026, has already taken interest rates to their highest since 1995, and much more is expected as Japan normalises its financial markets. India has roughly $50bn of yen-denominated external debt (6.4% of total). 

Each 100 bps of additional yen funding cost on a 30-year ¥1 trillion exposure is roughly ¥100bn in extra lifetime interest. This shifts the IRR calculus materially.

A JICA yen loan signed today has an all-in cost of about1.3–1.4%, still the cheapest money India can get, but roughly double the 0.6–0.8% that typical JICA loans carried in 2015–2022. More consequentially, as new BoJ hikes feed through, the TORF benchmark will keep rising. 

Put simply, India’s historical reliance on JICA for metro rail, DMIC, and large infrastructure is predicated on a yen funding cost that is quietly disappearing. This is not a cyclical shift, but structural one.

For the government, and including state governments, there is a need for a serious revisit of the external aided project (EAP) financing windows. Does it make fiscal sense to continue the same levels of such borrowings given the rate risks? 

Beyond the direct interest cost, higher global rates reset India’s external exposure through three further channels. They are the foreign portfolio flows into Indian debt markets, the yield that must be offered on NRI deposits to retain the stock, and the capital-wars channel through which US issuance crowds out EM access.

Finally, as we wrote in The Rise of Finance, thanks to the unavoidable spillovers from the dominance of the US economy and the dollar, the space available for developing countries to pursue autonomous monetary policy was already constrained. This is set to shrink further on the face of the trends discussed above. 

Weekend reading links

1. Thanks to inflation global debt burdens have been declining, falling by 27 percentage points between early 2021 and end of 2023. 

2. Rent controls may do more harm than good, except in controlled apartments.

And Spain is a good illustration that immigration is adversely impacting the housing market.

3. Excellent article on ASML, 42-year old, Veldhoven-based, Dutch maker of the $400 million extreme ultraviolet (EUV) lithography machines. 

The size of a double-decker bus, the machine prints circuits on to silicon wafers by firing light generated from 60,000 droplets of molten tin per second at them, using mirrors, lasers and short-wave light in a kind of vacuum-sealed nano-photocopier... ASML’s supply chain is something else. Where a typical industrial product might have a few dozen suppliers, an ASML EUV has more than 100,000 parts provided by 200 principal suppliers and 2,000 companies overall... Rather than trying to make everything itself, ASML outsources to the best manufacturer of each component — Zeiss for optics, Trumpf for lasers, VDL for mechatronics.

4. BigTech AI borrowings have fragmented the US debt market into two parts - AI-borrowing and the rest - and there is a growing risk that the cost of capital for everyone will rise as AI borrowings rise.

The scale of the borrowing will have its impact in squeezing out the rest of the economy and raising cost of capital for everyone. And it is going global.

In June, Amazon sold C$14bn ($10bn) of debt, smashing the Canadian currency’s previous record of C$8.5bn set by Alphabet just a few weeks earlier... Last month Alphabet issued a record A$5.5bn ($3.9bn) bond in Australia, more than twice as large as the previous biggest. The group also holds the record for the largest Swiss franc bond issue, having raised SFr3.1bn ($3.7bn) there in February. Amazon’s SFr2.8bn offering in May is the second-largest... Alphabet sold £5.5bn worth of sterling bonds in February, part of a $31.5bn package that included a rare 100-year £1bn bond that some say rekindled interest in long-maturity sterling debt. Amazon also raised £4.25bn earlier this month, its first-ever sterling issue.

What complicates matters is that the surge in AI borrowings come precisely at a time when government borrowings are also rising and are competing with each other, putting pressure on bond yields. 

5. India equity market facts of the week.

Both the Nifty and the Sensex... have ended in the red for eight straight weeks, their longest losing streak in 25 years. In the first nine months of 2026, foreign portfolio investors (FPIs) have pulled out at least ₹3 trillion from Indian equities, surpassing all annual outflow records.

6. Excellent chronicle by MS Sahoo of the emergence of India's regulatory architecture since 1991.

7. African credit ratings fact of the week.

Of an estimated $4tn regional capital base, less than a quarter is covered by credit ratings, according to the African Union. At the end of 2025, it said, the continent had fewer than 4,000 ratings, compared with 823,000 in the EU and more than 2mn in the US.

8. John Mearsheimer describes the Iran war the "mother of all strategic blunders".

Iran is number one in my opinion, because 1) we went into that war based on the false claim that Iran was on the verge of acquiring a nuclear weapon; 2) we had a theory of victory — air power alone would produce regime change — that has never worked and was not going to work this time; 3) we had no contingency plan for dealing with the likelihood that Iran would close the Strait of Hormuz; 4) we failed to achieve any of our four principal objectives — regime change, ending Iran’s nuclear enrichment capability, ending Iran’s long-range missile capability, and ending Iran’s support for Hamas, Hezbollah, and the Houthis; 5) the Strait was closed, which has already done significant damage to the international economy and could do much more — remember President Trump worries about following in Herbert Hoover’s footsteps; 6) we seriously undermined our carefully crafted alliance structure in the Gulf; 7) virtually all of our bases in the region have either been badly damaged or destroyed, to the point where we have moved most of our military assets out of the Gulf region and will probably not be able to return; 8) we have run down our stocks of key weapons to the point where we have weakened our deterrent postures in East Asia, Europe, and even the Western Hemisphere; 9) we have done enormous damage to our reputation as a reliable and wise ally; 10) we have markedly improved Iran’s strategic position in the Gulf region; and 11) there does not appear to be much chance — perhaps no chance — that we will get a nuclear deal of any sort with Iran, which was once considered of the utmost importance. In fact, there is a reasonable chance Iran will acquire nuclear weapons. President Trump has certainly given Iranian leaders — and other leaders of non-nuclear states around the world — a powerful incentive to acquire a nuclear deterrent.

9. Martin Wolf points to why the Iran war did not disrupt the oil markets and world economy as much as expected, with real oil prices still only around the long-term average. 

Johannes Urpelainen of the Johns Hopkins School of Advanced International Studies notes that: “First, the energy transition has reduced the global economy’s dependence on oil. Second, supply chains have proven resilient and adaptive. Third, producers outside the Middle East have acted fast to produce more oil and natural gas. Finally, drawdowns of global oil inventories have filled the remaining gap.” Flexibility is notably true of the Gulf. As former diplomat Mohammed Elsoukkary writes: “by the end of September preliminary estimates placed regional crude exports at approximately 16.3 million barrels per day, compared with 19.5 million in February, including roughly 9.7 million moving through Hormuz.”

Interestingly, the world entered the War with record inventories, accumulated especially in 2025.

10. French bond yields are higher than Italy's.

Not only does it now cost France more to borrow than it does Italy and Greece; a rising number of big French companies enjoy lower market interest rates than does the French state. Of all the G7 nations, it has suffered most from the debt sell-off triggered by the Iran war... France’s debt has reached €3.5tn, at nearly 120 per cent of GDP, far above the Eurozone average. Servicing costs have climbed fast to become the state’s single biggest expenditure ahead of education or defence. In just a month the additional interest rate that France pays to borrow over ultra-safe German debt — a key barometer of investor concern — has surged by almost two-thirds to around 1.4 percentage points on 10-year debt, heading towards the heights charted during the Eurozone debt crisis... A series of ratings downgrades has helped push up the country’s borrowing costs, which passed Greece’s in late 2024 and Italy’s last year. Investors now label France the “new periphery”... Despite the government’s tax-raising powers, its borrowing costs now exceed those of big French multinationals such as LVMH, the luxury group, and Sanofi, the pharmaceutical company. According to Goldman Sachs, more than a quarter of investment-grade French companies had a lower yield on their bonds than comparable government debt as of the end of September...

Increasing taxes to deal with the rising cost of debt servicing could also be perilous. At 44 per cent of GDP, the tax burden is already among the world’s highest; the average in the OECD club of rich countries is 34 per cent. Almost a quarter of public spending is gobbled up by pensions. French retirees leave the workforce earlier and receive relatively generous benefits by OECD standards. Attempts to push back the retirement age trigger angry protests.

11. US Treasury bonds are now threatening to hit 6% for the first time since the turn of the millennium. 

Borrowing costs for companies with the lowest credit ratings hit their highest level since May 2020 this month at 17 per cent, driven by the rise in Treasury yields to multiyear highs and by investors demanding more compensation for lending to such businesses. The risk premium for companies rated triple C or lower has risen to 12 percentage points, the biggest since 2022. Analysts said companies with floating-rate loans, and those needing to refinance debt in the coming months, were most at risk from the government bond sell-off, which last week pushed the yield on the 10-year Treasury note to its highest since 2002... But a further rise in Treasury yields could lead to broader pain across credit markets, analysts say. When it comes time to refinance existing debt, higher rates risk jarring the many companies that took on debt when interest rates were at rock-bottom levels in the early 2020s. Moody’s estimated last year that a record $1.45tn of US investment-grade corporate debt would come due between 2026 and 2030. Rising rates will put pressure on businesses to increase profits at a similar pace to rising borrowing costs.

12.  S&P quarterly earnings growth to remain above 25% for more than three quarters in a row. 

13. Ethnic cleansing in West Bank and East Jerusalem now. 

14. Interesting analysis of Adani Group's acquisitions through the NCLT process.

Since 2018, Adani Group, its linked entities and consortiums have participated in bids to acquire 26** distressed companies through India’s bankruptcy process. It has won 16 of them, with one case ongoing, though the latest decision went in Adani’s favour. This is a 61.5 percent success rate – meaning it won nearly three out of every five bids. While 14 of these 16 acquisitions came through the Corporate Insolvency Resolution Process (CIRP) to revive distressed business, the other two came through liquidation aimed at winding up operations... Across the 16 Adani acquisitions where tribunal orders disclose both admitted claims and recovery, creditors were owed approximately Rs 1.31 lakh crore and recovered only about Rs 33,000 crore – just 25.4 percent of the total. The loss was estimated to be over 97,700 crore, making the haircut around 74.6 percent. That is slightly higher than the national average of 69.4 percent.