The Existential Financial Gaps in Logistics (#72)

A deep dive into the market dynamics of Indian commercial fleets, and learnings from experiments on working capital, fuel, and commercial vehicle insurances

Jul 30, 2024

Welcome to the 72nd issue of Unit Economics. In this article, I share our team’s learnings from the last six months of conversations and experiments with commercial fleets, as we ambitiously imagined solving their $100 Bn working capital gap. If you are keen to learn about the logistics sector or have been looking at vertical fintech opportunities, you should find this particularly interesting. Dive in!

A Brief on Logistics and Road Transportation

India’s GDP is likely to have breached $4 trillion last financial year. The logistics sector alone makes up an estimated 8.5% of this GDP, much in line with the share of transportation in the US. The sector also provides livelihood to 22 Mn people, almost as much as the population of Australia.

Narrow down further and we see that Road Transport contributes to ~60% of the sector’s value, or individually to ~5% of the country’s GDP. This makes Road Transport as large as the Textile & Apparel industry and two-times the E-commerce industry in India today.

To internalise the extent of its involvement in our daily lives, imagine the supply chain for the last Amazon or Zepto order you made: your packaged goods would go from the point of production to a hub in its first mile, then processed again to another hub, before being transferred to the closest destination center, and eventually delivered to your doorstep.

In this process, commercial fleets power the entire supply chain from the first to the last mile. And there are more than 15 Mn vehicles on Indian roads today that do this daily (not accounting for the passenger three-wheelers or buses). Quietly underlining all the consumption, production, and trade in the country.

Over decades, many organised commercial fleets have attempted to capture large parts of the commercial road transport segment. Yet the largest of fleets continue to remain only a fraction (<0.01% each) of all commercial vehicles on the road. In fact, the sector remains so deeply fragmented that the small-fleet operators continue to make up 70% of all commercial fleets.

Let me share a quick primer on this.

70% of commercial fleets in India are run by Small-Fleet Operators (SFOs), i.e. with 5 or less vehicles. Almost half or more of these are DCOs (Driver-cum-Owners), i.e. where the driver owns and drives a single vehicle.
An estimated 20% of the commercial fleets are run by Medium-Fleet Operators (MFOs), i.e. between 6 to 20 vehicles. Most of these fleets have gradually progressed over the years from small to medium fleet size.
The remaining 10% are run by Large-Fleet Operators (LFOs), with 20 or more owned vehicles. Within LFOs, 90% of operators own or run less than 100 vehicles. Those with 100 or more vehicles are considered incredibly big in the sector and often dominate certain regions of operations.

Notice how ~90% of all commercial fleets continue to be small or medium in size. While these are undoubtedly critical to our supply chains, they today face market dynamics that pose an existential risk to their business. I explain three such headwinds below.

Increasing prices without cost-adjusted revenue

Commercial fleet business has traditionally been low-margin, with a net operating margin of 4-5% being the norm. Lately, this margin has been squeezed down further. This is because while the prices of diesel, driver salaries (supply shortage), and vehicle ownership (via EMIs) have increased, their revenue continues to be accrued in the broker markets on a fixed-contract basis.

Tax incentives discourage renting from small fleets

Small and medium-sized fleets are largely sub ₹10 crore turnover businesses and at this scale, they have incentives to keep themselves out of the burden of tax compliances. Moreover, a good part of the revenue - particularly for SFOs - is dependent on leasing their vehicles to large fleets.

However, with the implementation of GST and the increasing input tax credit incentives on the purchase of trucks, tyres, spare parts, etc., large fleets are today financially encouraged to buy rather than rent vehicles. This helps them realise a significant 4-5% additional profit margin.

For small and medium-sized fleets, this directly hits their top-line and forces them to (a) be more price-competitive and (b) to partner directly with businesses for loads. Else, they risk shutting down.

No access to working capital

Small fleets, like many other small Indian businesses, face the classic dilemma of positive cash-conversion cycles. Where their receivables follow a 30-90 day cycle, while the expenses on fuel, FASTag, salaries, tyres, insurance, commercial vehicle EMIs, permits, etc. are often payable within 0-14 days. These high up-front costs make sure that the fleets remain cash-poor on most days.

Today, they make do by taking 50-80% of their invoice value in advance. But it only hardly keeps them afloat, and reveals a large working capital gap.

Now, if the above three risks do materialise at scale, I anticipate that there isn’t nearly enough appetite or capital with the large fleets to replace the demand for loads (at least not at the speed of the churn of the small and medium-sized fleets). Neither do we find the next generation of smaller fleet operators optimistic of continuing on the same path.

A likely implication of this is a supply bottleneck, and consequently higher logistics costs and turnaround times for supply chains.

Our team spent the last six months trying to understand if we could bridge the third problem, i.e. the working capital gap, a gap larger than $100 Bn. We first attempted to tackle this head-first. I share what we learned and how it evolved in some detail below.

Banks and NBFCs have no option but to limit their commercial fleet exposure

Low margins and a high risk of mortality are often reasons enough to moderate working capital exposure to a sector. But in logistics, we notice that the problems run deeper:

Weak financials and lack of data: We saw lenders find consistently weak promoter and balance sheet records for even the large, old fleets. Particularly, we noticed two common trends: (a) balance sheets of fleet operators were already leveraged for capital expenditure and (b) revenue growth in the P&L was often inconsistent, given the external factors impacting the sector lately. Both factors lower the chances of an underwriting approval from the lenders. Worse, for the smaller fleets, we found none of this data to be readily accessible. And the informal data that was accessible was often manipulated and of low-trust.
Behavioural gaps in anchor recourse: Alternatively, to cover their underwriting risk, lenders seek some recourse or guarantee from borrowers. A common recourse is to only underwrite invoices of a well-regarded anchor. However, for this segment, we found the process of anchor evaluation to be of higher scrutiny than for other vendor segments. This is because, in times of stress, corporations de-prioritise the payouts of fleet operators first. We understood this to be due to the lower switching costs for corporates of their fleet vendors, compared to the other vendors.
Structural gaps in vehicle recourse: We also noticed that an average of 50% of existing vehicles for a fleet were already pledged with a financier for an existing CV loan. This leaves, on average, the rest of the 50% of vehicles as potential recourse to lenders. However, (a) the barriers to repossession of vehicles, (b) the low resale value due to old age of vehicles (average age today is as high as 9.5 years!), and (c) the high on-ground operational costs of doing so, make it extremely cumbersome for lenders to imagine depending on vehicle recourse for guarantee.

These factors have led to an environment where almost all lenders term logistics a “non-core” sector, i.e. of less focus and low exposure. Moreover, the perception of risk has only strengthened since COVID due to the higher default rates on new and used CV loans, segments that are otherwise relatively better understood by lenders. As a result, some lenders have even sunsetted the segment from their portfolios.

A few lenders, such as Sundaram Finance and Shriram Finance, continue to specialise in lending to commercial fleets. Saying that, their books remain largely dependent on new vehicle loans, with little exposure to working capital credit.

Eventually, with all the known variables, we realised that a direct attempt at tackling the working capital issue leaves lenders and us with an unfavourable risk-reward tradeoff. If we were to solve the problem, we needed to build more confidence to underwrite by accessing richer data.

The big data disadvantage in working with Indian commercial fleets

Now, on the subject of data, there has been some notable progress in the digitisation of commercial fleets over the last decade. Let me paint a picture of how it has changed.

A decade back, if you were to look into how small or medium-sized commercial fleets make payments, you would notice a reliance on cash for >90% of all business expenses. There were a few loyalty cards, but only with very large fleets. There was no UPI. There were no large-scale mobile applications or logistics-specific ERP systems to run operations. Even the smartphone penetration with small and medium-sized fleet operators was less than 50%. All of their business happened on phone calls, paper notes, and handwritten invoices.

Today, a lot has changed. We notice that each fleet operator uses at least five separate mobile apps to run their operations: one each for GPS, FASTag, Fuel, UPI transfers, and Messaging. Interestingly, the adoption for each of these has been through force. For context,

In 2016, the Automotive Industry Standards Committee (AISC) established guidelines for the implementation of Intelligent Transport Systems (ITS), which made Vehicle Location Tracking (VLT) mandatory for all motor vehicles. This immediately pushed up the adoption of GPS and paved the way for companies like WheelsEye, Fleetx, and LocoNav.
The same year, Jio made the internet cheap, demonetisation forced a temporary move away from cash, and the newly introduced UPI and FASTag started picking up in small pockets. These developments led to the adoption of WhatsApp messaging, Fuel loyalty cards, and FASTag in droves for fleets.
In 2020, COVID hit. And that made UPI payments primary for all commercial fleets. This had a more defining impact, with cash payments continuing to be less than 40% of all fleet payments today, versus 90% or more a decade back.
The next year, the National Highways Authority of India (NHAI) mandated the use of FASTag for all toll payments, digitising what are ~5% of all business expenses for commercial fleets transporting goods between cities.

You would think that all of this would led to a welcome change in the attitude towards technology for commercial fleet operators. To certain extent, this is true. But while the progress is promising, to access any such data, the ground realities are less so:

While GPS or rather VLT was enforced on fleet operators, small and medium-sized fleets continue to see them with very little utility and are not comfortable with paying much for the hardware or any additional fleet management use-cases. This has (a) created an extremely crowded market, where players sell GPS knock-offs from China at cheap, almost unsustainable prices, and (b) the quality of data captured via the devices is not very rich. This is reflected in the US-first strategy shift for many Indian telematics providers. To make matters worse, there are strong reservations from providers to commercialise their GPS data through APIs. Unavailability of vehicle tracking or other fleet management data limits a lot of use cases that have benefited fleets elsewhere (e.g. managing payments fraud). The lack of data also leads to mis-pricing of many key financial assets, particularly related to insurance and credit, which could ideally be tailored more appropriately.
FASTag payments data, usually mapped with the toll location, amount, and balances, can also prove to be a rich source - yet faces the same data availability issues. Further, the barriers to entry for FASTag programs have increased significantly today with a 31% decrease in interchange, and a One Vehicle, One FASTag policy, which makes the replacement of existing FASTags high-friction.
Thirdly, a great alternative to acquiring reliable fleet data is via enabling UPI transactions. This would require one to become a Third Party Application Provider (TPAP) for the use-case of fleet expenses. This is probable, yet doing so requires an understanding that: (a) with a relatively low scale, the approval for a TPAP licence from NPCI is likely to be a lengthy process, (b) the infrastructure and maintenance costs of a TPAP program will make this a capital-intensive GTM, and (c) the acquisition of fleets through this mode will require a cash-back incentive to force a switch in behaviour, which is likely to again require high-capital investment.

What else could we look at? Maybe fuel payments. How about a fuel card to save fleets money? A topic explored by many logistics aggregators in India before, not least because of the billion-dollar outcomes elsewhere (Wex, Fleetcor [now Corpay], and AtoB).

We spent a good time (a) understanding the pains and behaviours around fuel payments, which led to experimenting with a fuel wallet, and (b) eventually around commercial insurances. And our learnings there bring me to the third realisation.

Poor incentives and anti-competitive markets in logistics need government intervention

On Fuel Payments

Fuel is the largest expense item for a commercial fleet. It makes up anywhere between 30% to 70% of a fleet’s running costs. As a rule of thumb: the longer the distance a commercial vehicle travels per delivery, the higher the proportion of its fuel expenses. Cumulatively, an estimated $60 Bn was spent by commercial vehicles in India on diesel fuelling last year. More than what Indians spent on smartphones.

Now, given the share of fuel in fleet expenses, if you could save 1% or more on these payments for a fleet, you (a) add meaningful margins to their otherwise low-margin business and (b) capture high mind-share by being the go-to payment provider for a daily use-case. An ideal GTM for a fintech focused on commercial fleets.

Unfortunately, and we learned this the hard way, the market dynamics make this infeasible. I will elaborate.

In India, 97% of all diesel fuelling happen at pumps of three Oil Marketing Companies (OMCs): Indian Oil, Bharat Petroleum, and Hindustan Petroleum.
The diesel prices, while deregulated, are still largely dictated by the Ministry of Petroleum given the PSU status of the three OMCs.
The price regulations have today led to a situation where the OMC earns a net ₹2 per litre of diesel it services. With higher margins on petrol, each OMC earns blended ~4-5% in net margins.

Now, given the government control on diesel prices through the OMCs and the low margins they earn on diesel fuelings, the PSUs have been able to hand-twist banks to agree to a zero MDR. For issuers, this implies that the costs of enabling payments at fuel pumps need to be recouped via surcharges to the consumer, i.e. the fleet operator, instead.

So, unlike the US counterparts, where the interchange on fuel payments allows favourable unit economics for tens of private co-issuers, that incentive is entirely taken away from open-loop debit or credit cards in India.

To a private entity that wants to save fuel expenses for fleet operators and earn some revenue while doing so, there is only one option: run a closed-loop program in direct partnership with OMCs and agree to a revenue-sharing that works for all sides.

However, while possible - and we attempted the same via a wallet with HPCL, your proposition here will initially be tied to the fixed rewards program of the OMCs, which is also dictated by the Ministry of Petroleum. As it so happens, the loyalty programs of all three OMCs offer a fixed discount or cashback of 0.4% today.

With this knowledge, we found that the only way to build a superior proposition for fleets on fuel payments is to negotiate rates directly with hundreds of individual dealers on the ground. And offer those in addition to the 0.4%. This is possible, in theory - yet it is an extremely hard and operational task in practice, as we later realised, especially given the low revenue share we would get from the OMCs or dealers.

On Commercial Vehicle Insurances

As a last resort, we went deeper to understand pain points in another critical financial instrument for fleets: commercial vehicle insurances. We did not imagine it to provide us rich data for underwriting, but we could hear fleet operators ask for lower premiums and a better claims experience. And with the large industry size for such insurances, we could see some promise.

Note: Third-party insurance has long been mandated for motor vehicles in India, and the Ministry of Road Transport and Highways guides the base premium for the same for all vehicle categories each year.

I will try to capture what we learned in brief below:

The base premium on goods carrying commercial vehicles ranges anywhere from ₹14K to ₹38K ($170-450) per year today. At an average premium of ₹20K across 15 Mn vehicles, the GWP of the market is sizeable at ₹30,000 crore (~$4 Bn). Add the Own Damage / First Party (OD) insurance, and the market is even larger.
Now, there are three key actors in CV insurance: General Insurer (Licensed underwriter & issuer of the policy), Direct Broker (General), and the Sub-Agent (end-distributor). We evaluated the first two, where a fintech would get to play more critical roles.
For a general insurer, the commercial vehicle portfolio is one of the riskiest, with the net margins negative or in small single digits. Our conversations with insurers helped us understand why it is so: (a) there is an incredibly high cost of settlements for third-party claims on life loss, and (b) there is rampant lobbying by the fleet operators who, along with the police and garage owners, manipulate the value of claims to realise higher value than the actual loss. These are two local behaviours that you can exercise very little control over.
As a result, while the general insurers are mandated to meet growth criteria set for commercial insurance GWPs each year, we noticed an active attempt by Business Heads of these organisations to limit the portfolio exposure to particular fleet types and RTOs. Due to the inherent market risks of commercial vehicle insurances, we also saw that the market leaders in GWPs are often also loss-leaders.
For a direct broker, commercial vehicle insurance is a lucrative business, with their share of premium as high as 35-40% (and without any underwriting risk). However, four market forces to consider here: (a) since the claims process is mostly out of a broker’s control, there is very little one can do to notably improve the claims experience for fleets from what is already a market standard today; (b) therefore, as a broker, your proposition to a fleet operator has to be largely around pricing, especially given the high cost of premiums; you will also find some appetite to switch without price discounts, if you can allow fleet operators flexibility to make premium payments on EMIs, although this would be a riskier endeavour; (c) long-haul and large CVs often operate from truck terminals, and at truck terminal, you will notice participants of local welfare associations [1] act as insurance agents for multiple partner brokers, and [2] have deep existing ties with fleet operators; creating a dent here would be challenging and will need on-ground relationship building. Lastly, (d) the high competition between brokers has pushed cash-discounts on premiums for third-party insurances to as high as 20-25% and on own-damage (OD) coverage to 50-99%, undercutting even the base premium, and leaving brokers with net premiums as low as 5%. This is, both, a P&L and a regulatory risk for brokers, especially when you want to build a large book.

A wish list and a few parting thoughts

The Government of India (GoI) has been quite vocal about its plans to boost efficiency and lower logistics costs across the country, introducing large-scale initiatives including PM GatiShakti, National Logistics Policy (NLP), Unified Logistics Interface Platform (ULIP), and the Logistics Data Bank. Given how unorganised the sector continues to be, the focus bodes well.

However, if the above experience suggests anything, it is that the free market appears to be driving the sector in an opposite direction, one of higher logistics costs and lower efficiency. And all of this is tied to the existential risk to the long tail of commercial fleets in the sector.

If the government are to wrest this control, it needs to intervene to create a more fair ground for smaller fleet operators. This requires a sharp focus on improving their unit economics, including but not limited to the following: (a) more regulated cost-adjusted freight rates in broker networks, (b) allocate shipping tenders towards associations of small fleets to increase their vehicle utilisation, (c) subsidies on fuel or other fleet expenses for fleets below a certain size or recorded turnover, (d) a managed marketplace to hire skilled drivers to resolve supply shortage, and importantly, (e) easier access to working capital.

I firmly believe that the large-scale changes, unfortunately, will only have to come from the government. The private entities, despite the intent, do not have sufficient monetary incentives to charitably enable small and medium-sized fleets to sustain better.

For instance, BlackBuck, catering mainly to SFOs with three or fewer vehicles, is doomed to serve a segment that has the lowest margins, highest mortality rate, and the least loyalty. Similarly, WheelsEye will struggle to justify the venture scale, with the small revenue pools in GPS, FASTag, etc. and the high risks in lending.

Many others share similar pains and will continue to - unless the government intervenes.

P.S. We eventually chose not to go ahead further with this particular area of work, since we could neither build confidence in the unit economics, nor find potential for solving the working capital problem effectively.

If you want to share your thoughts or chat on the topic, or if there are any other ideas you want to soundboard, please reach out on email or linkedin. Until next time!

Unit Economics