Peer Comparison in an IPO

Turn to the Basis for the Issue Price section of any Indian offer document and you will find a table headed something like Comparison with Listed Industry Peers. A handful of already-listed companies are listed in rows, with columns for revenue, earnings per share, price-to-earnings ratio, return on net worth and net asset value. The issuer's own figures sit alongside them. The table looks like a neutral piece of reference data. It is not.

Nobody outside the transaction picks those companies. The issuer selects them, in consultation with the lead managers, and the selection is disclosed as being made by the issuer. That is a legitimate and disclosed arrangement, not a hidden one. But it means the table is an argument presented in the form of data, and the most useful thing a reader can do is treat it that way.

What the table is meant to do

A price on its own is uninterpretable. Nobody can say whether 25 times earnings is a lot without knowing what similar businesses trade at. The peer table exists to supply that reference frame, so an investor can see the issuer's multiples next to multiples the market is already paying for comparable companies. Used honestly, it is a genuinely useful disclosure and one of the few places where a prospectus invites external comparison at all.

The catch is arithmetic rather than ethical. If the reference frame moves, the same price looks different without a single figure about the issuer changing. Place a company asking 30 times earnings next to peers trading at 45 times and the price reads as restrained. Place the identical company next to peers trading at 18 times and it reads as demanding. The issuer's numbers were constant in both cases. Only the neighbours changed.

What each column is actually measuring

The columns are fairly standard across offer documents, and each answers a narrow question. Knowing what each one covers, and more importantly what it leaves out, stops you from reading more into a row than it contains.

• Revenue from operations sizes the business, but a large top line says nothing about whether that revenue is profitable or whether it is concentrated in a few customers.
• Earnings per share divides profit by share count. It is sensitive to how many shares exist, so a company that has recently issued stock will show a lower figure for the same profit.
• Price to earnings compares the market price of a peer, or the offer price of the issuer, against that earnings per share. For peers it reflects a price the market sets continuously; for the issuer it reflects a price the seller is proposing.
• Return on net worth measures profit against shareholders' funds. A high figure can reflect genuine efficiency or simply a small equity base propped up by debt, so it should always be read next to leverage.
• Net asset value per share reports book value per share from the balance sheet. It matters a great deal for lenders and asset-heavy businesses, and much less for companies whose value sits in brands, contracts or people.
• The period covered, usually the most recent completed financial year, applies to every row and quietly determines the whole picture if that year was unusual for anyone in the table.

Notice the asymmetry buried in the price-to-earnings column. The peers' ratios are computed from prices the market has arrived at through years of trading, disclosure and scrutiny. The issuer's ratio is computed from a price that has never been tested by a market at all. Placing them in adjacent rows of the same table makes them look like the same kind of number, and they are not.

How peers actually get chosen

There is no exhaustive formula. The general expectation is that peers should be listed companies in a broadly similar line of business, of a broadly comparable scale, with disclosed and comparable financials. Within that expectation a great deal of discretion survives, and each of the following decisions is a judgement call the issuer makes.

• How the business is defined. A company can describe itself narrowly, by a single product line, or broadly, as a diversified player, and each description points to a different peer universe.
• Where the size boundary sits. Including much larger, established companies changes the picture; so does restricting the set to companies of similar revenue.
• Which listed companies count as comparable at all. In some industries only a few peers exist, and in others dozens do, which gives more room to select.
• Whether a peer whose business overlaps only partially is included or left out.
• Which financial year is used, and whether all rows are drawn from the same period on the same accounting basis.

The offer document is normally explicit that the peers have been identified by the company and that the selection is based on stated criteria. Read that note. It is short, it is easy to skip, and it tells you the table is a curated set rather than an exhaustive industry survey.

The ways a peer set can flatter without misstating anything

It is worth being precise here. The figures printed in the table are drawn from published accounts and are checkable. The framing effect comes from selection and context, not from wrong numbers. Below are the recurring patterns that change how a price reads.

1. Peers drawn from a segment with structurally higher multiples. A company that mostly does contract manufacturing but presents itself alongside branded consumer companies borrows a valuation frame it may not have earned.
2. A set weighted toward market leaders. Comparing a small regional operator against the two largest national companies in the sector can make the issuer look inexpensive, even though scale, distribution and bargaining power are not remotely similar.
3. Omission of the closest comparable. If an obvious, similarly sized competitor is absent, that absence is information. Perhaps it trades at a lower multiple, or perhaps there is a legitimate reason such as a very different business mix.
4. Very few rows. A table with two peers gives a thin basis for any comparison, and a single unusual peer can pull the perceived industry multiple a long way.
5. Mixed periods or bases. If the issuer's figures are restated consolidated and a peer's are standalone, or the years differ, the columns do not line up even though they appear to.
6. Ratios computed on different share counts. As covered in the valuation guide, a pre-issue multiple placed next to peers' post-issue market multiples is not a like-for-like comparison.

None of these patterns proves anything improper. A narrow industry may genuinely have only two listed peers. A company may have real reasons to see itself as adjacent to a premium segment. The point is that each of these is a claim about comparability, and claims can be tested.

An illustration of how much the frame matters

Consider an invented company that reports 60 crore rupees of profit on 6 crore post-issue shares, giving earnings per share of 10 rupees, and is offered at a cap of 300 rupees, which is 30 times earnings. Suppose the peer table shows three companies at 42, 48 and 51 times, averaging 47. The issuer appears to be asking for a discount of roughly a third to its industry.

Now imagine the same table with a different but equally defensible peer set: three companies at 19, 22 and 26 times, averaging roughly 22. The issuer at 30 times now appears to be asking for a premium of about a third. Every number about the issuer is unchanged. All the movement came from the reference group. Both these peer sets and all these figures are invented purely to demonstrate the effect.

This is why the peer table should never be the last step in forming a view on price. It is a starting frame, and frames can be replaced.

Building your own comparison

The most effective response to a curated table is to construct your own. This takes an evening, not a career, and everything needed is publicly available from exchange filings and company annual reports.

1. Write down, in your own words, what the company actually sells and to whom. Use the business section of the offer document, not the marketing language of the qualitative factors.
2. List the listed companies whose revenue comes from a similar activity, including any obvious competitor missing from the issuer's table.
3. For each, note revenue, profit, margins, return on net worth and debt from the latest published financial statements, keeping the period and the basis consistent.
4. Compute the multiples yourself from current market prices rather than importing numbers from the table.
5. Check the revenue mix of each peer. A company with 80 per cent of revenue from a different segment is a weak comparison whatever its headline classification.
6. Compare growth trajectories over several years, not one year, because a multiple mostly reflects expected growth.
7. Ask which peer the issuer most resembles in economics, not in description, and weight that one most heavily.

Step five carries more weight than its position suggests. Two companies in the same industry classification can have entirely different economics if one sells to a handful of institutional buyers and the other to millions of consumers. Customer concentration, pricing power and the working-capital cycle separate businesses far more sharply than an industry label does.

Things worth comparing beyond the multiple

The prescribed columns in the table are narrow, and the more revealing comparisons often sit outside them. When you build your own version, add rows for the things that explain why one company deserves a different multiple from another.

• Operating margin and its trend over several years, which tells you about pricing power more directly than any single-year ratio.
• Debt relative to equity and to operating profit, since a highly leveraged company and an unlevered one are not comparable on earnings multiples alone.
• Cash conversion: whether reported profit is showing up as operating cash flow.
• Revenue concentration by customer, product or geography.
• Capital intensity, meaning how much investment is required to add a rupee of revenue.
• Governance and related-party dealings, which do not appear in any ratio but influence how much investors are willing to pay.

Our guide on analysing IPO financials goes through where each of these is disclosed in the restated statements, and the risk factors section frequently states concentration and dependency issues more bluntly than anywhere else in the document.

Where the comparison genuinely breaks down

Sometimes there is no honest peer set, and it is better to acknowledge that than to force one. A company with a business model that has no listed equivalent in India cannot be framed by domestic multiples in any meaningful way. Offer documents in that position may say that no comparable listed peers exist, or may reach for international companies, which brings in different regulation, currency, market structure and cost of capital.

A loss-making company presents a related problem: earnings multiples cannot be computed, so the table shifts to revenue multiples or other measures, and the comparison becomes correspondingly looser. In these cases the honest conclusion is that the peer table carries less weight, and the analysis has to lean harder on the business itself, the unit economics and the path to profitability described in the document.

The table starts ageing the day it is printed

A peer table is a photograph, not a live feed. The peers' share prices move every session, which means their price-to-earnings ratios move too, while the number printed in the document stays frozen at whatever it was on the reference date stated in the section. Between the draft document and the final one, and again between the final one and the bidding period, the industry multiples the table relies on may have shifted noticeably in either direction.

The peer set itself can also change between the draft red herring prospectus and the red herring prospectus, as can the financial year used once fresh annual results become available. That is another reason to read the version of the document that actually governs the issue you are considering, and to recompute the peers' multiples from current prices rather than accepting the printed ones. Our guides on the DRHP and the RHP explain how the two documents relate.

A short reading discipline

Read the peer table twice. The first time, read it as presented and note the impression it creates. The second time, ask three questions: who is missing, are the periods and bases consistent, and would a different but equally defensible set change the conclusion. If the answer to the third question is yes, the table has told you less than it appeared to.

Then use it for what it is good at. A peer table efficiently shows you which listed companies the market considers adjacent, roughly what multiples that neighbourhood carries, and where the issuer's margins and returns sit relative to established players. That is genuinely useful context, provided you supply the scrutiny that the format does not.

As with everything in an offer document, work from the latest filed version. Peer sets and figures can change between the draft and the final document, and market multiples for the peers move continuously. Verify company data against exchange filings and the companies' own published statements before drawing any conclusion.