Relative Valuation (Comps)

First published 23 Aug 2026 · Last verified 29 Aug 2026

Lesson 47.1 — The Neighbourhood Comparison: What Relative Valuation Really Means

Imagine you are trying to figure out a fair asking price for a house in Baneshwor. You would not compare it to a bungalow in Budhanilkantha with three times the land area, nor to an apartment in a completely different city. You would look for houses of similar size, similar age, similar road access, sold in the last few months, in the same or a neighbouring ward. If those comparable houses sold for roughly NPR 2.5 crore, and your house is similar in every important respect, then NPR 2.5 crore becomes your anchor. You might adjust up or down for a slightly bigger plot, a busier road, or a better view — but the starting point is what similar things actually sold for.

That is the entire logic of relative valuation, sometimes called "comps" (short for comparables). Instead of building a valuation from scratch by projecting every future cash flow of a company for the next twenty years and discounting it back to today — the approach you will study in the discounted cash flow (DCF) chapters later in this Part — relative valuation asks a simpler question: what is the market currently paying for businesses similar to this one, and does that tell us something useful about what this business should be worth?

You have already built the tools for this in Chapters 39 through 41, where you learned to compute and interpret the core ratios: the price-to-earnings ratio (P/E, the share price divided by earnings per share, telling you how many years of current profit you are paying for), the price-to-book ratio (P/B, the share price divided by book value per share, telling you how much you are paying for each rupee of accounting net worth), EV/EBITDA (enterprise value divided by earnings before interest, tax, depreciation and amortisation, a measure that looks at the whole company including its debt, not just the equity slice), and the dividend yield (the annual dividend per share divided by the share price, telling you the cash return you receive simply for holding the stock). Relative valuation is what you do with those ratios once you have them: you stop looking at a single company's multiple in isolation and start asking, "compared to what?"

There are, in fact, two different kinds of "comparable" that this chapter will use side by side, and it is worth separating them clearly from the start because Nepali investors often blur them together.

The first is cross-sectional comparison: looking across several companies at the same point in time. If Nabil Bank trades at a P/B of 2.1x and Global IME Bank trades at a P/B of 1.3x, and both are large, well-run commercial banks with broadly similar business models, that gap is cross-sectional information. It tells you something about how the market is pricing one bank relative to another, right now.

The second is time-series comparison: looking at one company against its own history. If Chilime Hydropower Company has historically traded at an average P/E of 18x over the last five years, and it is currently trading at 30x, that gap tells you something different — that the market's view of Chilime today is unusually optimistic (or pessimistic) compared to its own recent past. This does not automatically mean the stock is overvalued; something may have genuinely changed (a new plant coming online, a tariff revision, a change in interest rates that makes utility-like cash flows more attractive). But it flags a divergence worth investigating.

A seasoned analyst uses both lenses together. A bank trading in line with its peers but far above its own five-year average multiple is telling you a different story than a bank trading far below its peers but in line with its own history. Confusing the two — treating "cheap relative to history" as the same thing as "cheap relative to peers" — is one of the quiet errors that creeps into otherwise careful analysis.

It helps to be explicit about why relative valuation is popular, and why this book treats it as a complement to, not a replacement for, intrinsic valuation methods like discounted cash flow. Relative valuation is fast: you can screen twenty-seven commercial banks on P/E and P/B in an afternoon, using data readily available from NEPSE, Sharesansar, or Merolagani, without building a full financial model for each one. It reflects the market's current mood, which matters because you eventually have to sell your shares to that same market, not to a spreadsheet. And it is intuitive: "this bank is cheaper than that bank" is a sentence any investor, professional or retail, can understand immediately.

But relative valuation has a structural weakness baked into its logic, and it is worth stating plainly before you go any further: it assumes the peer group is priced correctly on average. If every commercial bank in Nepal is simultaneously overpriced because retail money has flooded into NEPSE during a bull run, then a bank trading "cheap relative to its peers" may simply be the least overpriced stock in an overpriced sector — not a genuinely attractively priced asset in absolute terms. Relative valuation tells you about relative position, not absolute worth. Keep that sentence in mind; it will resurface throughout this chapter, and it is the reason Lesson 47.5 introduces the idea of a "justified multiple" derived from fundamentals rather than simply borrowed from the crowd.

Lesson 47.2 — Building the Right Peer Group on a Concentrated Exchange

Choosing comparable houses in a real estate search is easy because there are thousands of houses to choose from in most cities. Choosing a comparable set of listed companies on NEPSE is much harder, because the exchange is small, concentrated, and lopsided in ways that most textbook treatments of relative valuation — usually written with the New York Stock Exchange or Bombay Stock Exchange in mind, where there are dozens or hundreds of companies in any given industry — do not prepare you for.

As of 2026, NEPSE lists roughly 300 companies, but the distribution across sectors is extremely uneven. Banking and financial institutions (commercial banks, development banks, finance companies, and microfinance institutions) together with the hydropower sector account for the large majority of listed companies and a substantial share of total market capitalisation. Insurance is a meaningful third group. Manufacturing, trading, hotels, and "others" make up the remainder, often with only a handful of listed names in each. This is the opposite of a market like the S&P 500, where you can find twenty or thirty reasonably similar companies inside a narrow sub-industry. On NEPSE, your "peer group" for a mid-sized commercial bank might be all 17-20 other commercial banks — a wide set — while your peer group for a niche business (say, the single listed cigarette manufacturer, or the single listed cement producer of a particular type) might be one company, or none at all.

This has a direct practical consequence: peer group construction on NEPSE requires more judgment and more caution than the same exercise would on a larger exchange, precisely because your options are narrower and the temptation to stretch the definition of "comparable" to fill out a table is stronger.

Start peer group selection with the criterion you already learned to respect in Chapter 39: sector-specific structural differences matter more than surface-level similarity. A peer group should be built around companies that share the same fundamental economic engine, not simply the same stock exchange or the same size of market capitalisation. For NEPSE, the two dominant clusters deserve separate treatment:

Commercial banks should be compared to other commercial banks — and ideally to banks of similar scale and business mix. Nepal has large, well-capitalised banks such as Nabil Bank, Global IME Bank, Nepal Investment Mega Bank, and NIC Asia Bank, which sit at a different scale than smaller commercial banks. Within banking, you should also separate commercial banks from development banks and finance companies, since these operate under different regulatory tiers (the "A," "B," and "C" class institutions under Nepal Rastra Bank's classification), with different minimum capital requirements, different permitted activities, and different risk profiles. A "B" class development bank and an "A" class commercial bank both take deposits and make loans, but comparing their P/B ratios directly, without adjusting for scale and regulatory tier, misses real structural differences.

Hydropower companies should be compared to other hydropower companies — but even within hydropower, you must further separate operating companies from pre-operational or construction-stage companies. A hydropower company that has been generating and selling electricity for ten years, such as Chilime Hydropower Company or Butwal Power Company, has an earnings stream you can measure today. A hydropower company still under construction, with its first unit not yet commissioned, has no meaningful current earnings at all — its P/E ratio, if it has one, is often a meaningless or even negative number, and investors are really pricing a claim on future cash flows that do not yet exist. Comparing the P/E of an operating hydropower company to a pre-operational one is comparing a fruit tree to a sapling: both may become valuable, but only one is bearing fruit right now.

Beyond sector and sub-sector, a disciplined peer group in the Nepali context should also account for:

Scale. A commercial bank with a loan book of NPR 200 billion faces different economies of scale, different funding costs, and different regulatory scrutiny than one with a loan book of NPR 40 billion. Larger banks often (though not always) command a premium multiple because of perceived stability, deposit franchise strength, and liquidity of their shares.

Ownership structure and promoter concentration. Many NEPSE-listed companies, especially hydropower companies, have promoter shareholding locked in for a statutory period (a topic covered in more detail in Lesson 47.4), which affects how much of the company's shares actually trade freely. Two hydropower companies with identical plants and identical revenue can have very different observed multiples simply because one has a much smaller tradeable float than the other.

Growth stage and capacity utilisation. A hydropower company that just commissioned a new, larger project and is still ramping up toward full capacity utilisation will show rapidly growing earnings for a few years, even without any change in the underlying asset. Comparing its P/E to a mature company running at stable, flat output conflates "temporarily depressed current earnings due to ramp-up" with "genuinely lower value."

Capital structure. Hydropower projects in Nepal are typically financed with a high proportion of debt relative to equity — often in the range of 70:30 or 80:20 debt-to-equity during the construction and early operating years, financed through syndicated loans from Nepali banks and development finance institutions. Two companies with similar plant sizes but very different leverage will have very different EV/EBITDA-to-P/E relationships, and comparing their P/E ratios alone, without looking at EV/EBITDA (which neutralises the effect of the capital structure by looking at the whole enterprise), can be misleading. This point is developed further in Lesson 47.6.

A practical peer group exercise, then, looks less like "find five NEPSE companies with a market cap near mine" and more like a checklist: same core sector, same regulatory tier, comparable scale, comparable operating stage (mature and generating stable revenue, versus growth or pre-operational), and — where the differences cannot be eliminated by selection — adjustments applied explicitly rather than ignored.

Lesson 47.3 — Why You Cannot Compare a Bank to a Hydropower Company

It is worth spelling out, in concrete terms, exactly why cross-sector comparison on NEPSE is dangerous — not as an abstract warning, but by walking through what would actually go wrong if you tried it.

Suppose an investor notices that a mid-sized commercial bank trades at a P/E of 12x, while a well-known hydropower company trades at a P/E of 35x, and concludes that the bank is "cheap" and the hydropower company is "expensive," recommending a switch from hydropower into banking stock. This reasoning treats P/E as if it means the same thing in both cases. It does not.

A commercial bank's earnings are driven by net interest income (the spread between what it earns on loans and investments and what it pays on deposits and borrowings), fee income, and loan loss provisioning, all of which move with the credit cycle, interest rate policy set by Nepal Rastra Bank, and deposit competition among the roughly twenty commercial banks. A bank's earnings are typically less volatile year to year than a hydropower company's, because a diversified loan book smooths out the ups and downs of individual borrowers, but a bank carries credit risk — the possibility that borrowers default — that a hydropower company largely does not carry in the same form.

A hydropower company's earnings, by contrast, depend on river flow (which varies seasonally and year to year — Nepali rivers run high in monsoon and low in winter, creating a "dry season" earnings dip that recurs every single year), the power purchase agreement (PPA) it has signed with the Nepal Electricity Authority (NEA) or another off-taker (which fixes the tariff it receives, often with a wet-season and dry-season rate, and often without meaningful escalation for many years), and its debt service burden, since most hydropower projects are financed with large loans that must be repaid regardless of how much water flows. A hydropower company's earnings can also look artificially small, or even negative, immediately after a new plant is commissioned, because depreciation and interest expense are front-loaded while revenue is still ramping toward full-capacity output — meaning a "high" P/E is not necessarily "expensive," it may just reflect earnings that are temporarily depressed relative to the company's normalised future earning power.

These are not minor stylistic differences; they are structurally different businesses with different margins, different growth trajectories, different risk factors, and different relationships between current accounting earnings and true underlying value. A P/E of 12x for a bank, whose earnings are relatively stable and recurring, is not directly comparable to a P/E of 35x for a hydropower company whose earnings are still ramping up toward a normalised run-rate. The hydropower company's P/E might fall to 15x within three years purely from capacity ramp-up, with no change in share price at all, simply because the "E" in P/E grows. The bank's P/E of 12x might already reflect a fairly mature, fully-provisioned earnings stream unlikely to grow much faster than nominal GDP.

The same logic applies, with less obvious but equally real force, within sectors that look similar on the surface. Life insurance companies and non-life (general) insurance companies are both "insurance," but they have different reserving requirements, different claim patterns, and different capital regulations under the Nepal Insurance Authority; comparing their P/B ratios one-for-one without adjustment is a milder version of the same mistake. Microfinance institutions and commercial banks both lend money, but microfinance serves a different borrower base, carries different credit risk, operates under a different regulatory ceiling on interest rate spreads, and often trades at systematically different multiples because of these structural realities — not because the market is randomly mispricing one relative to the other.

A useful mental discipline here is to ask, before comparing any two multiples: "If I owned 100% of both of these companies outright, with no stock market involved, would a rupee of reported earnings, book value, or EBITDA mean roughly the same thing in both?" If the answer is no — because one company's earnings are far more volatile, one carries far more financial or operational risk, one is growing much faster, or one operates in a fundamentally different regulatory box — then the two multiples are not directly comparable, and any conclusion drawn from comparing them raw is unreliable.

Lesson 47.4 — When the Market Itself Distorts the Multiple: Illiquidity and Thin Float on NEPSE

Every method of relative valuation rests on one quiet assumption: that the observed market price is a reasonably reliable signal, produced by enough genuine buying and selling interest that it reflects a real consensus about value. On a deep, liquid market with millions of shares changing hands daily across a broad ownership base, that assumption mostly holds. On NEPSE, for a meaningful number of listed companies, it does not — and an analyst who forgets this will read false signals directly out of the data.

Start with the structural reason this happens. Under the Companies Act and securities regulations enforced by the Securities Board of Nepal (SEBON), and further shaped by hydropower-specific rules, promoter shareholders — typically the founders, sponsoring institutions, and early strategic investors — are required to hold their shares for a statutory lock-in period after listing, commonly several years, before they are permitted to sell into the market. This is meant to signal commitment and prevent founders from dumping shares immediately after an IPO. But it has a side effect directly relevant to relative valuation: the free float (the portion of total shares outstanding that is actually available to trade in the open market, as opposed to being locked up with promoters, the government, or other long-term holders) can be a small fraction of total shares outstanding — sometimes well under 30% for a newly listed hydropower company, versus a much larger free float for an established commercial bank with a broad, long-listed shareholder base.

A small free float matters enormously for relative valuation because it directly affects how reliable the observed price — and therefore the observed multiple — actually is. When only a small percentage of shares are tradeable, a relatively modest amount of buying or selling can move the price sharply, simply because there are not enough shares available to absorb the order without a large price change. NEPSE additionally applies daily circuit breakers (price bands limiting how much a stock can move in a single trading session), which, for a thinly traded stock, can mean it takes many consecutive trading days of hitting the upper circuit for the price to reach a level that reflects genuine aggregate demand — and during that entire climb, or during a subsequent slide, the multiple you observe on any given day is a snapshot of a price still finding its level, not a settled market judgment.

This produces several concrete distortions worth naming explicitly:

Newly listed, small-float hydropower stocks often trade at extremely high multiples in their first months or years on the exchange — not necessarily because the market has done careful fundamental work and concluded the projects are worth that much, but because retail demand for a small number of freely tradeable shares outstrips supply. An investor who takes such a multiple at face value and uses it as a peer benchmark for valuing another, similarly small hydropower company is anchoring on noise, not signal.

Even for more established, moderately liquid stocks, low daily trading volumes mean the "last traded price" on a given day may reflect a handful of transactions, sometimes involving related or connected parties, rather than a broad market clearing process. A multiple calculated from that day's closing price can swing meaningfully based on a single large trade.

Some multiples are simply unstable at the level of significant digits when volume is thin: a P/E that jumps from 22x to 27x over one week is more likely to reflect illiquidity-driven price noise than a genuine 23% reassessment of the company's earning power.

A related distortion worth flagging is the effect of NEPSE's broad market cycles on sector-wide multiples. Nepal's stock market has historically moved through pronounced bull and bear phases, often driven by retail sentiment, margin lending availability, and macro liquidity conditions (interest rates, remittance inflows, banking sector liquidity) more than by company-specific fundamentals. During a broad rally, average P/E and P/B ratios across an entire sector — banking, hydropower, or otherwise — can rise together, well above levels historical fundamentals would justify, simply because more money is chasing the same shares. In such an environment, using "the sector average multiple" as your benchmark for what a company "should" trade at risks anchoring an entire valuation exercise to a temporarily inflated crowd consensus. This is precisely the failure mode flagged at the end of Lesson 47.1, and it is the reason the next lesson introduces a check that does not depend on what the crowd happens to be paying today.

Lesson 47.5 — The Justified Multiple: Deriving a Fair Multiple from Fundamentals

Everything covered so far in this chapter compares a company's multiple to what other companies, or the same company in the past, are trading at. That approach has an obvious blind spot: if the whole peer group, or the whole market, is mispriced, comparing within that mispriced group tells you nothing about whether the group itself is cheap or expensive in any absolute sense. The tool that addresses this blind spot is the justified multiple — a multiple derived not from what the market happens to be paying today, but from the company's own fundamentals: its profitability, its growth prospects, and the return investors require for bearing its risk.

The clearest and most widely used version of this idea, particularly well suited to banks (where book value is a meaningful, closely tracked figure because balance sheets are the core of the business), is the justified price-to-book ratio. It is derived directly from the dividend discount model applied to a company growing at a constant rate — the same Gordon growth logic you will see formalised in the DCF chapters later in this Part, but expressed here in ratio form rather than as a discounted cash flow. The formula is:

Justified P/B = (ROE − g) / (Ke − g)

Where ROE is the company's sustainable return on equity (net income divided by shareholders' equity, the profitability metric you studied in Chapter 40), g is the expected long-run growth rate of earnings and book value, and Ke is the cost of equity — the annual return that shareholders require for holding this specific company's risk, typically estimated using a model such as the Capital Asset Pricing Model (CAPM), built up from a risk-free rate (commonly proxied in Nepal by long-term government bond or development bond yields), an equity risk premium, and a beta reflecting the stock's volatility relative to the market.

The intuition behind this formula is worth sitting with, because it explains something every Nepali bank-stock investor has noticed but perhaps not been able to articulate: why do some banks trade at a P/B of 2.5x while others, seemingly similar in size, trade at 0.9x? The formula says the answer lies in the gap between ROE and cost of equity. A bank earning an ROE well above its cost of equity is creating value with every additional rupee of equity capital it deploys — book value compounds and the market is willing to pay more than one rupee of price for one rupee of book value, precisely because that rupee inside the bank is working harder than the market's required return. A bank earning an ROE close to, or below, its cost of equity is not creating economic value with new capital, even if it is "profitable" in an accounting sense — and the market rationally refuses to pay a premium over book value for it, sometimes even pricing it below book value (a P/B under 1.0x).

This single insight reframes what "cheap" and "expensive" mean in a way that pure cross-sectional comparison cannot. Two banks might both trade at a P/B of 1.5x — identical on the surface — yet one might be fully justified by its ROE and growth profile while the other is significantly overpriced relative to its fundamentals, simply because its ROE is lower or its risk (and therefore its cost of equity) is higher. A peer comparison alone would tell you these two banks are "priced the same." A justified multiple calculation tells you they are not equally attractive at all.

The same logic extends, with appropriate adaptation, to other multiples covered in Chapters 39-41:

Justified P/E, under the same constant-growth framework, can be expressed as the dividend payout ratio divided by (Ke − g). This is particularly relevant for hydropower companies, many of which have policies of paying out a large share of earnings as dividends once they reach a stable operating phase, since they have limited need to retain capital for new growth beyond their licensed capacity.

Justified EV/EBITDA is harder to express in a single clean formula because it depends on capital structure and reinvestment assumptions, but the underlying principle is identical: a company's fair EV/EBITDA should rise with its EBITDA margin, its growth in EBITDA, and the durability of its cash flow (a long-dated, government-counterparty PPA supporting cash flow for another twenty-five years justifies a different multiple than a PPA expiring in three years), and should fall as its risk and cost of capital rise.

Dividend yield can be checked against a "required yield" benchmark derived from the cost of equity minus expected capital growth — if a stock's dividend yield sits well below what its risk profile would suggest investors should demand, absent strong growth to compensate, that gap is itself a signal worth investigating, echoing the "high-yield trap versus growth" distinctions you studied when dividend yield was first introduced in Chapter 41.

It is worth being honest about the limits of this tool as well. The constant-growth assumption behind the justified P/B and justified P/E formulas is a simplification; it assumes ROE, growth, and cost of equity are all stable over a long horizon, which is rarely exactly true, especially for a hydropower company whose earnings profile changes materially as it moves from ramp-up to full capacity utilisation, or for a bank whose ROE can be temporarily depressed by a large one-off loan loss provision. The formula is a lens for organising your thinking and locating the drivers of a valuation gap, not a precise, mechanically "correct" answer to plug into a spreadsheet and trust blindly. Treat the justified multiple the way a doctor treats a diagnostic test: it points you toward the right question, it does not replace the full examination.

Lesson 47.6 — Common Pitfalls and a Disciplined Comps Checklist

With the mechanics of peer selection, sector-specific danger, illiquidity distortion, and justified multiples now covered, this final lesson consolidates the recurring mistakes that undermine relative valuation in practice on NEPSE, and closes with a disciplined process you can apply every time.

Pitfall one: comparing a growth story to a mature one without adjustment. A hydropower company that just commissioned a new run-of-river plant and is still ramping toward full capacity, or a bank rapidly expanding its branch network and loan book in underserved provinces, will show rising earnings for reasons that have nothing to do with the market reassessing its multiple. Comparing its current P/E directly to a mature, slow-growing peer's P/E without adjusting for this growth difference — or better, without normalising both companies' earnings to a comparable point in their respective life cycles — systematically misreads growth companies as "expensive" and mature companies as "cheap," when in fact a higher multiple may be entirely justified by a higher growth rate, exactly as the justified P/E and justified P/B formulas from Lesson 47.5 would predict.

Pitfall two: ignoring differing capital structures. As Lesson 47.2 noted, hydropower projects are frequently financed with a high proportion of debt. Because interest expense sits below the operating line, two companies with identical plant economics and identical EBITDA can show very different net income, and therefore very different P/E ratios, purely because one carries more debt than the other. A highly levered company's P/E can look deceptively low precisely because a larger share of its enterprise value is financed by debt rather than equity, concentrating both the upside and the risk onto a smaller equity base. This is exactly why EV/EBITDA — which is calculated using enterprise value (market value of equity plus net debt) rather than just equity market value, and EBITDA rather than net income — is often the more reliable multiple for comparing capital-intensive, debt-financed businesses like hydropower companies, since it neutralises the distortion that different leverage levels introduce into P/E. When comparing companies with meaningfully different debt-to-equity ratios, always cross-check P/E-based conclusions against EV/EBITDA before drawing a conclusion about relative cheapness.

Pitfall three: relying on stale or manipulated earnings. Nepali companies, like companies everywhere, do not always report earnings that cleanly reflect sustainable, repeatable operating performance. A bank that books a large one-off gain from selling investment securities, or that under-provisions for loan losses in a given quarter to flatter reported profit, will show a temporarily inflated EPS and therefore an artificially low, falsely attractive P/E. A hydropower company reporting earnings for a year with unusually strong river flow (a wet year) will show a P/E that looks cheap relative to its own more typical, average-hydrology years — precisely the seasonal and annual variability discussed in Lesson 47.3 — and an investor who anchors on that one favourable year's earnings without normalising for average hydrological conditions across a full cycle will overestimate the sustainable earning power the multiple is being paid for.

Pitfall four: treating a peer-group average as a target price. Even after carefully building a clean, well-matched peer group, it is tempting to conclude that if the peer average P/B is 1.6x and your target company trades at 1.2x, the "fair" price is simply the peer average — full stop. This skips the justified multiple step from Lesson 47.5 entirely. The correct sequence is: identify the peer group, observe the peer average, then ask whether your target company's own ROE, growth, and risk genuinely support trading at that peer average, above it, or below it. A company with a below-peer-average ROE deserves a below-peer-average multiple; concluding it is "undervalued" simply because it trades below the peer average, without checking whether its fundamentals justify a discount in the first place, is a common and costly error.

Pitfall five: forgetting the effect of thin liquidity discussed in Lesson 47.4 when constructing a peer average. Including one or two illiquid, small-float names in a peer average can pull the whole benchmark toward a distorted level, especially in a small peer set (recall from Lesson 47.2 that NEPSE peer groups are often narrow to begin with, sometimes only five to ten names). A single thinly traded hydropower stock with an inflated multiple, sitting inside an eight-company peer group, can move the simple average meaningfully; a float-weighted or liquidity-screened average is more robust.

To bring these pieces together, consider two illustrative peer comparison tables, of the kind an analyst would build before making any relative valuation judgment on NEPSE. These figures are illustrative and rounded for teaching purposes, constructed to reflect realistic relationships between scale, profitability, and multiples rather than to serve as live market data — always pull current figures from NEPSE, Sharesansar, or company disclosures before acting on any real comparison.

Table 1: Illustrative Peer Comparison — Nepali Commercial Banks

Reading this table the way this chapter has taught you to: Nabil Bank's P/B of 2.1x sits well above the peer average of 1.6x, but its ROE of 15.2% also sits above the peer average of 13.2% — some or all of the premium multiple may be justified by superior profitability, which is exactly the kind of question the justified P/B formula from Lesson 47.5 is built to answer, rather than simply flagging Nabil as "the expensive one." Himalayan Bank, by contrast, combines a below-average ROE (10.5%) with a below-average P/B (1.2x) — here, the discount may be entirely fundamentals-driven rather than a hidden bargain, a distinction the peer average alone cannot make for you.

Table 2: Illustrative Peer Comparison — Nepali Hydropower Companies

Notice how this table makes visible several of the pitfalls this chapter has warned against. Upper Tamakoshi's P/E of 38x looks dramatically more expensive than Chilime's 16x on a raw comparison — but Upper Tamakoshi is still ramping up toward full-capacity earnings, exactly the growth-versus-mature distortion flagged in Lesson 47.6's first pitfall, and its EV/EBITDA of 11.0x is far closer to the mature peers' 9.0-9.5x than its P/E suggests, because EBITDA is less distorted by ramp-up depreciation and interest patterns than net income is. The illustrative newly listed, small-float project shows the most extreme multiples in the table on every measure — precisely the thin-liquidity and small-float distortion described in Lesson 47.4 — and a disciplined analyst would either exclude it from a peer average entirely or flag it as unreliable rather than treating its 85x P/E as a meaningful benchmark for anything.