Ask most NEPSE investors why they bought a hydropower stock, and you will hear a number: megawatts. The project is 25 MW, or 100 MW, and bigger sounds better. Almost nobody can tell you what rate the company will be paid for each unit of electricity, for how many years, or whether the buyer is even obligated to pay if it cannot use the power. That information is not a secret. It sits in one document, the power purchase agreement, and this is the hydropower PPA explained in the only terms that matter to a shareholder: the PPA, not the megawatts, is the business.
The PPA is the contract between the project company and the Nepal Electricity Authority (NEA), the near-monopoly buyer of grid power in Nepal. It fixes the price, the volume, the duration, and the conditions under which money actually changes hands. A turbine spins, but the PPA is what turns spinning into revenue. Two projects with identical capacity, identical rivers, and identical engineering can be worth very different amounts purely because their PPAs are written differently. If you are going to value a hydropower stock at all, you have to read the offtake terms. This piece is the deep-dive companion to our broader framework on how to value a hydropower stock on NEPSE.
The tariff: two prices, not one
The first thing to understand is that a Nepali run-of-river project does not get paid a single rate. It gets paid two, split by season, and the gap between them is large.
NEA’s standard PPA pays a higher rate in the dry season and a lower rate in the wet season. The logic is straightforward: in the monsoon, rivers are swollen and the country is awash in power, so each unit is worth less; in winter, flows drop, supply tightens, and energy is scarce, so each unit is worth more. The historically quoted figures for run-of-river projects were around NPR 4.80 per unit in the wet season and NPR 8.40 per unit in the dry season. The seasonal split itself is also a design choice the regulator has changed over time. Reporting indicates the dry-season window was extended from four months to six (running roughly December to May, with the wet season June to November).
Why does this matter for your numbers? Because a run-of-river plant generates the most electricity exactly when the price is lowest. Heavy monsoon flow produces big volumes paid at the low wet-season rate. Lean winter flow produces small volumes paid at the high dry-season rate. The blended price you actually earn depends on the plant’s generation profile across the year, not on either headline rate alone. This is the same seasonality that whipsaws reported earnings quarter to quarter, which we unpack in hydropower seasonality and earnings. It is also why the difference between run-of-river and reservoir projects is partly a PPA story: a storage or peaking plant that can hold water and release it in the dry season captures more of the high rate, and NEA has paid higher dry-season tariffs for peaking projects to reward exactly that.
One more tariff wrinkle for large projects. For schemes above 100 MW, NEA has signed PPAs partly denominated in US dollars, but only for a portion of the term, typically tied to the foreign loan repayment period, after which payment reverts to rupees. For a shareholder, a dollar-linked tranche is a hedge against rupee depreciation for those early years and ordinary rupee exposure afterwards.
Escalation: a fixed number of raises, then flat forever
Here is the clause that quietly caps a project’s lifetime revenue, and it is the one investors most often get wrong.
The PPA includes an annual escalation, meaning the per-unit rate rises by a set percentage each year. The catch is that it does not rise forever. The standard structure has been an escalation of around 3 percent per year, applied a fixed number of times, commonly eight, from the start of commercial generation, and then the tariff stays flat for the rest of the contract. Projects such as Rasuwagadhi and Mid-Bhotekoshi were reported on exactly this basis: roughly 3 percent annual escalation up to eight times from the date commercial production begins.
Read that carefully, because it is not inflation protection. After the escalation steps are exhausted, the rate is frozen in nominal rupee terms. If general inflation runs at, say, 6 to 7 percent a year, the real value of that frozen tariff erodes for the entire remaining life of the PPA, which could be two decades. A project commissioned today might see its rate climb for eight years and then sit unchanged until the 2050s while costs around it keep rising.
Work a quick example. Suppose a unit starts at NPR 4.80 in the wet season. Escalated at 3 percent for eight years, it reaches roughly NPR 6.08, an increase of about 27 percent over eight years. After year eight, it stays at NPR 6.08 for the remainder of the contract. The headline often sold to investors is “the tariff goes up every year.” The reality is “the tariff goes up for eight years, then never again.” Those are completely different businesses when you discount the cash flows.
The term: shorter than the dam
The contract term is the number of years NEA is bound to buy the power. In Nepal this has typically been in the range of 25 to 30 years from the commercial operation date, with 30 years common for larger schemes. Negotiation and project size move the exact figure.
The point investors miss is that the PPA term is usually shorter than the physical and licensed life of the project. A well-built hydropower plant can run for fifty years or more, and generation licenses can extend beyond the PPA. So what happens to the cash flows in the gap between the PPA expiring and the plant retiring? Honestly, nobody can promise you. The power could be sold to NEA under a renewed agreement at a then-unknown rate, sold to private buyers if a market exists by then, or the asset could be handed over to the state depending on the license terms. When you value a hydropower stock, the safe approach is to value the contracted PPA years with some confidence and treat everything after expiry as a speculative tail, not a guaranteed annuity. A 30-year PPA on a 50-year asset is not a 50-year cash machine.
Take-or-pay versus take-and-pay: where the real risk lives
Everything above sets the price and the timeline. This last clause decides whether the price and timeline mean anything at all. It is the most important sentence in the document, and it comes down to two words.
Under a take-or-pay PPA, NEA must pay for the contracted energy whether or not it actually takes it onto the grid. If the transmission line is down, if demand is low, if the system simply cannot evacuate the power, NEA still pays. This is also called deemed generation: the plant is treated as having delivered, and it is compensated for what it would have earned. Take-or-pay is what makes a project bankable. Banks lend against a PPA because it produces a predictable stream of revenue largely independent of NEA’s day-to-day grid problems. The risk of NEA failing to take the power sits with NEA, the buyer, not with the developer or its lenders.
Under a take-and-pay PPA, NEA pays only for the electricity it actually takes. If the grid cannot absorb the output, the plant spills the water, generates nothing it gets paid for, and earns nothing for that period. The risk of evacuation, transmission constraints, and seasonal oversupply shifts onto the developer. The turbines can be in perfect condition and the river full, and the revenue can still be zero on a day NEA chooses not to buy.
For an investor, the distinction is existential. A take-or-pay contract is close to a bond: contracted volume, known price, obligated buyer. A take-and-pay contract is closer to selling into a spot market controlled by a single customer who decides how much to buy. The same plant, the same tariff, and the same term produce a fundamentally different risk profile depending on which of those two phrases appears in the offtake clause. This is precisely the kind of clause you should be hunting for when you read an IPO prospectus in Nepal for a hydropower IPO on NEPSE, and almost nobody does.
The 2025 budget shock
This is not theoretical. In Nepal it nearly became the new default overnight.
The annual program and budget for fiscal year 2025-26 carried a line stating that PPAs for run-of-river projects would be done on a take-and-pay basis. According to the Independent Power Producers’ Association, Nepal (IPPAN), that single provision threatened to close the path for nearly 17,000 MW of projects because banks and financial institutions do not lend against take-and-pay PPAs. IPPAN warned the shift jeopardized roughly NPR 109 billion already invested in ongoing projects and put a far larger pipeline at risk. The reaction across the sector was severe enough that the Kathmandu Post described the energy sector as “rocked” by the policy, and developers framed it as an attempt to push the private sector out of hydropower.
The episode is the cleanest proof of the argument in this piece. Nothing about the rivers, the turbines, or the megawatts changed in 2025. What changed was two words in the offtake clause, and the market treated it as a threat to over a hundred billion rupees of value. The government later signaled the provision had been removed from the budget speech, and committees were formed to study whether future PPAs should be take-or-pay or take-and-pay, leaving the question unsettled rather than closed. For investors, the lesson is permanent even if the policy is not: the offtake clause is a live political variable, and it can move.
How to actually read it
You do not need to be a lawyer. When you look at a hydropower company’s prospectus or annual report, find the PPA summary and answer five questions in order.
First, what are the wet-season and dry-season tariffs, and what is the realistic blended rate given the plant’s generation profile? Second, what is the escalation, how many times does it apply, and what is the flat rate afterwards? Third, how many years is the term, and how does that compare with the asset’s life? Fourth, and most important, is it take-or-pay or take-and-pay, and is there a deemed-generation clause that pays the company when NEA cannot evacuate the power? Fifth, is any part of the tariff dollar-denominated, and for how long?
If a prospectus is vague on the fourth question, treat that as a red flag, not a detail. The macro backdrop matters too, since NEA’s ability to honor take-or-pay obligations depends on its own finances and on the interest rate and liquidity environment shaped by NRB monetary policy. A take-or-pay promise is only as strong as the entity making it.
The verdict
Megawatts are the number that gets quoted. The PPA is the number that gets paid. A larger plant with a take-and-pay contract, a frozen post-escalation tariff, and a term well short of its asset life can be a worse investment than a smaller plant locked into take-or-pay with healthy seasonal rates. The capacity tells you how much electricity the project can make. The PPA tells you how much money the company will keep, for how long, and under what conditions, and the offtake clause tells you whether that money is contractually owed or merely hoped for.
Read the agreement before you read the brochure. In Nepali hydropower, the contract is the company.
This is analysis, not financial advice.
