How to Analyse Group Structure and Minority Interests

Reading a group is a procedure, not an impression. Six steps run in order: establish which statements are in hand, map the structure from the notes, locate where the profit and the assets actually sit, separate what the parent's shareholders have no claim on, list what is participated in but never consolidated, and write down what would settle each remaining question. The output is a corrected picture with its gaps named.

Before the first step, what the procedure produces at the end is worth fixing in mind. The procedure produces a set of figures that mean what they appear to mean, a map of who sits where, and a short written list of questions with the document that answers each one. The procedure produces no view on whether the structure under examination is sensible, tidy, aggressive or evasive. The refusal to produce that view is not modesty and it is not caution. A set of accounts carries no evidence about why a structure was built. Any sentence about intent is therefore an invention wearing the clothes of arithmetic.

Four things are taken as already settled here and none of them is rebuilt. The first is the test that decides whether one company is consolidated into another at all. The second is the mechanics of consolidation itself, the adding across and the cancelling out that turn two sets of books into one. The third is how goodwill arises and what its impairment signals. The fourth is what an associate and a joint venture are and how influence without control is accounted for. Each is set out in its own right elsewhere, and all four are used here as inputs rather than derived again. The work that remains is reading a set of group accounts in a fixed order, saying where each figure came from, separating the part of the group's profit the parent's shareholders can actually claim, and recognising the one step whose omission is the commonest arithmetic error in the whole subject.

In what order is a group examined, and why does the order matter?

The order is not a preference, and each position in it is earned. Which statements are in hand comes first because the words assets, profit and earnings per share carry different meanings on either side of that answer, and a reading built on the wrong side cannot be repaired further down. The group structureThe list of companies that report together as one, together with how much of each one the parent company holds and how each is brought into the reported figures. is mapped second because arithmetic performed over a membership list that was never written out is arithmetic performed over a guess. Nothing belonging to other people can be taken out until it is known which company earned it, so locating the profit precedes adjusting it. A total is exactly the wrong instrument for finding something it was never asked to include, so the participations sitting outside the reported totals are listed before the reading stops. The questions come last for the simplest reason available: they are what the procedure produces.

The six steps are a sequence rather than a menu, and the step readers skip most often, separating the profit that belongs to holders outside the group, is the only one of the six whose omission puts a wrong number rather than a missing number into the answer. The figure below rewards two passes. The first follows the numbers down the spine and nothing else. The second ignores all six and looks only at the red panel underneath. The three additions in that panel are not advanced work, and they are not what comes next once the six become easy. All three fall outside the procedure altogether, and a reading that wanders into any of them has quietly changed into something else.

Step one: which set of statements is actually in hand?

The fault this step prevents has nothing to do with accounting, and every reader has lived through it. A man says his household spends Rs 46,000 a month. A month later, the same man says Rs 31,000. Nothing changed in the household. The first figure counted everybody living under the roof, including his brother's share of the rent and food; the second counted only what he and his wife pay from their own salaries. Both figures are honest, both are useful, and anybody who compares one against the other has compared two different households and learned nothing about either.

Step one asks two questions and computes nothing at all: are these standalone figures or consolidated figures, and if consolidated, which company sits at the top of them. Both answers get written down before a number is touched. A set of accounts usually carries both, printed one after the other, and the two are not variations of a theme. The two sets answer different questions about different things, and a single figure lifted from the wrong one will run through every later step without ever announcing itself.

Run it on Anjani Stationers Private Limited, an invented business that makes school notebooks and exercise books and holds 70 per cent of Chitra Binding Works, also invented, bought at the start of year two. Both sets are available. The standalone set shows assets of Rs 1,80,00,000, profit after tax of Rs 30,00,000 and basic earnings per share of Rs 7.50. The consolidated set shows assets of Rs 2,09,50,000 and profit after tax of Rs 40,00,000. Same business, same year, two completely different sets of figures, and the difference is not an adjustment or a restatement. The gap is the answer to a different question. And there is a third set nobody prints on the cover. Chitra Binding Works has its own standalone position: assets of Rs 47,00,000, liabilities of Rs 12,00,000 and net assets of Rs 35,00,000. The third set matters at step three, so note now that it exists.

Step two: which notes give the structure?

Mapping a group is clerical work, and it is the step people most often skip. The front of a set of accounts gives an impression of the structure, and the impression feels sufficient. It is not. The map comes from three specific notes, read together, and each one gives something the other two do not.

Step two builds a written list of every member, the holding in each, and the treatment applied to each, taken from the subsidiaries note, the associates and joint ventures note, and the basis of consolidationThe note in a set of group accounts that states which companies have been included, on what footing each one was brought in, and where a conclusion differs from what the holding percentage alone would suggest. note. The subsidiaries note gives the names and the holdings. The associates and joint ventures note gives everything the group participates in without consolidating. And the basis of consolidation note is where any conclusion that differs from the arithmetic of a percentage is stated in words. A percentage on its own can mislead, and the basis note is the one place that says so. The whole list gets built before anything is computed with it.

Anjani Stationers takes about four minutes. The subsidiaries note carries one line: Chitra Binding Works, 70 per cent, acquired at the start of year two, consolidated in full. The business participates in nothing else, so the associates and joint ventures note is empty. The basis of consolidation note states that the whole of Chitra Binding Works is added in and that the share belonging to the other holders is shown separately. One line is the entire map. The one line gets written down anyway, in that form. A one line map written down is a map, and a one line map carried in the head is an assumption that stops being tested by step four.

Step three: where inside the group does the profit actually sit?

Ten shops trading under one signboard report one takings figure at the end of the month. The takings figure says how the signboard did. The takings figure does not say that eight of the ten covered their rent and no more, that one lost money quietly all month, and that the tenth, the one beside the bus stand, carried the whole result. The month's total is completely true and it hides the only thing worth knowing about the month.

Step three asks which member of the group generated the profit and which member holds the assets, and the answer changes what every other figure in the statements means. Two sources supply it. The first is the segment noteThe note that breaks a group's revenue, result and assets into the separate lines of business or geographies its management actually runs and reviews, rather than reporting only one combined total.. The segment note splits the group's figures by the parts management actually runs. The second is any summarised information given for individual members. Where neither exists, subtraction gets a long way: the difference between the consolidated figures and the parent's own standalone figures is the rest of the group.

Applied to Anjani Stationers, the answer arrives in one line of arithmetic. Consolidated profit after tax is Rs 40,00,000 and the parent's standalone profit after tax is Rs 30,00,000, so Rs 10,00,000 of the group's profit was earned inside Chitra Binding Works. Rs 10,00,000 is a quarter of the group's profit, exactly 25 per cent, sitting in a company the parent does not hold in full. The same subtraction on the balance sheet gives Chitra Binding Works assets of Rs 47,00,000 out of the group's Rs 2,09,50,000. The share of assets is 22.43 per cent. So both the profit and the assets tell the same story: a little over a fifth of this group sits in a company where somebody else has a claim. Nothing has been adjusted yet. Where things are has simply been found out. Finding out is the whole of step three.

Step four: what changes when what the parent's shareholders cannot claim is separated?

Four cousins buy a shop together. One of them puts in seven rupees of every ten and the other three put in the rest between them. The shop makes Rs 1,00,000 in a year. The cousin who put in seven tenths can say truthfully that the shop she runs made a lakh, and she can say truthfully that Rs 70,000 of it is hers. Both sentences are correct. Only one of them says what she got.

Step four separates the consolidated figures into the part attributable to ownersThe share of a group's profit or equity that belongs to the shareholders of the parent company, as distinct from the share belonging to the outside holders of a partly held subsidiary. of the parent and the part belonging to the non-controlling interestThe holders of the shares in a subsidiary that the parent does not hold. Their share of the subsidiary's net assets and profit is reported separately inside the group figures. Older writing and everyday conversation still call this the minority interest., and when the question is what the parent's shareholders have a claim on, the owners' figures are the ones that answer it. Both figures are printed. Nothing is being derived here; the work is choosing correctly between two lines that already sit on the face of the statement, one under the other. The older name for the second line is the minority interest. Both names still appear constantly, in conversation and in older writing, and both are worth recognising.

On Anjani Stationers the separation is exact. Of the group's Rs 40,00,000 of profit after tax, Rs 37,00,000 is attributable to the owners of Anjani Stationers and Rs 3,00,000 belongs to the outside holders of Chitra Binding Works, being their 30 per cent of the Rs 10,00,000 Chitra earned. Now carry the split into earnings per share. Earnings per share is where the choice actually bites. Anjani Stationers has 4,00,000 shares. Rs 37,00,000 over 4,00,000 shares is Rs 9.25. Not Rs 10.00, the figure the whole Rs 40,00,000 would give, and not the standalone Rs 7.50 either. Three different earnings per share figures are defensible from this one group in this one year, Rs 7.50, Rs 9.25 and Rs 10.00, and only the middle one says what a share in Anjani Stationers earned. The same split runs down the balance sheet: of the group's equity of Rs 1,59,50,000, the owners hold Rs 1,49,00,000 and the non-controlling interest is Rs 10,50,000.

Step five: what is participated in but never consolidated?

A woman runs a tailoring unit of her own and also puts money into a cousin's fabric shop, where she has a third of the shares and no say in how it is run. Her own books show the tailoring unit in full and the fabric shop as one line. If the fabric shop takes a loan and she has signed for it, nothing in her books gets bigger, and she is still the person the lender will call. Everything she participates in is not everything her books add up.

Step five lists what the group participates in without consolidating, because a group can be materially larger in participationTaking part in the economics of another business, through a shareholding, an arrangement or an obligation, whether or not that business is added into the reported totals. than it is in reported figures, and because obligations given to entities outside the consolidated totals are real obligations that consolidated liabilities do not carry. Three things go on this list. Associates and joint ventures, whose assets and borrowings sit outside the consolidated totals however large they are. Any guarantee or commitment given to something outside the group. And any arrangement that gives the group a share of results without giving it a company to consolidate. Where an associate exists, the summarised financial informationA short set of figures given in the notes for an associate or joint venture, typically its assets, liabilities, revenue and profit, so a reader can see the size of something that was never added into the group totals. in the notes is what states its size.

On Anjani Stationers this step produces a short answer and one item worth care. There are no associates and no joint ventures, so nothing at all sits outside the consolidated boundary. Anjani Stationers carries a guarantee of Rs 8,00,000 over borrowing taken by Chitra Binding Works, disclosed in its standalone notes and exactly the shape of item this step exists to catch, except that here it points inward: Chitra is consolidated, so Chitra's borrowing is already inside the group's liabilities of Rs 50,00,000, and the guarantee adds no exposure the consolidated figures have missed. Had that guarantee been given to an associate instead, the borrowing would have sat outside those liabilities entirely and the guarantee would have been the only trace of it. Write down that the slot is empty. An empty slot checked is worth more than a full slot assumed.

Step six: what gets written down, and where does the work end?

Of the six, this is the one that feels least like analysis and travels furthest. Everything the earlier steps could not close becomes an entry, and every entry carries the name of the paper that would close it. A request for better disclosure in general is not an entry. Something a person could ask for by name, and receive, is.

The rule for stopping has no judgement in it: once the statements are established, the structure is mapped, the profit is located, the owners' share is separated and the participations are listed, the reading is finished, and it finishes with live questions still open. A document answers those questions, not another pass over the same figures. Past that line the procedure is no longer running at all. Anything that continues past the line is estimation, and the estimate arrives dressed in the credibility of five steps that had nothing to do with it.

Four entries close out Anjani Stationers, each with its paper named. Whether Chitra Binding Works can declare a dividend without its outside holders agreeing, answered by its articles and by any shareholders' agreement between the parties. What terms govern the trading between the two companies, given Rs 8,00,000 of binding work invoiced by Chitra during year two with Rs 1,50,000 of it outstanding at the year end, answered by the related party note and the intercompany agreements. What Chitra borrows in its own name and on what conditions, answered by its own filed accounts. And what event would trigger a call under the Rs 8,00,000 guarantee, answered by the guarantee document. The four entries are what the reading produces, they are the correct place for it to stop, and handing them over completes the work rather than abandoning it.

The three findings that matter are written out here in full. At the default holding of 70 per cent, step three finds Rs 10,00,000 of the group's Rs 40,00,000 earned inside Chitra Binding Works, step four separates Rs 37,00,000 attributable to owners from Rs 3,00,000 belonging to the outside holders, and earnings per share comes out at Rs 9.25 against the Rs 10.00 the skipped route reports, an overstatement of 8.11 per cent. Lowering the holding to 60 per cent gives the same group Rs 36,00,000 attributable, earnings per share of Rs 9.00, and an overstatement of 11.11 per cent. At 51 per cent it is Rs 35,10,000, Rs 8.775 a share and 13.96 per cent. The subsidiary's profit never moves across that whole range and the group total never moves either, yet the figure a shareholder in the parent can actually claim falls the whole way. Step four is therefore a step and not a refinement.

How does the sequence read when all six positions are set down together?

All six steps sit below with their outputs on Anjani Stationers and no prose between them. Take the right hand column on its own and count how little of it is calculation: two rows carry arithmetic, three carry a list, and the bottom row carries no figure whatever.

Not one amount in that table was estimated: each is printed, disclosed, or noted as absent on purpose, and the closing row holds no amount at all. A properly completed reading takes exactly that shape. What tempts a reader at the foot of that column is a seventh row holding an opinion. The table has six rows.

Who actually reads a group this way, and what do they do afterwards?

In any given week four different readers work through some version of these six steps, and the divergence lies entirely in what happens to the entry list afterwards.

A lender works through it to identify which company in the group its money is actually lent to, an equity analyst to secure the right numerator, a holder inside a partly held subsidiary to learn what the majority can settle without asking, and Vaidehi Rao to have the answers ready before the questions arrive. Begin with the lender, whose use of it is the least obvious and carries the heaviest consequence. A lender advancing money to Anjani Stationers has a claim on Anjani Stationers, and the assets of Chitra Binding Works are not Anjani Stationers' assets. Chitra Binding Works has its own creditors, and they rank ahead of anybody lending upstairs. So the lender's step three question is not how large the group is; it is how much of the group sits in the company whose name is on the loan document. On these figures, Rs 47,00,000 of assets sits in the other company, and the lender's step six list will ask for Chitra Binding Works' own accounts by name.

The equity analyst uses the output more narrowly and more often. Every multiple, every growth rate and every margin they build has a numerator, and step four decides it. Using Rs 40,00,000 where Rs 37,00,000 belongs is not a rounding difference, and it repeats in every year and every forecast built off that year. The investor in Chitra Binding Works has the opposite reading of the same map: they hold 30 per cent of a company whose customer, whose finance and possibly whose pricing all sit with the majority holder, so their step six question about dividends is the one that decides whether their share of Rs 10,00,000 ever reaches them. And Vaidehi Rao, finance controller at Anjani Stationers, gets more direct use from the four entries than anyone. The four entries are the exact things a lender or an analyst will raise with her, and every one of them can be closed out of paper already sitting in her cupboard. Holding the paper is the everyday advantage of standing inside a business, and it is never available to anybody reading the same accounts from outside.

Which additions to this sequence are always wrong?

A reader who finds four written entries a thin reward for an hour's work tends to reach for one of three additions, and none of the three can be carried by a set of accounts. Judging whether a structure suits the business is the first. Reading intent out of complexity is the second. Treating an outside holding as a signal about something is the third.

Take the second one seriously. Reading intent out of complexity is the addition that feels like insight. A group with fourteen subsidiaries across four countries looks, to a reader who has just learned this procedure, like something. A structure of that size is not something. Businesses end up structured the way they are for reasons that leave no trace in the accounts at all. A company was bought and kept as a separate company because unwinding it would have cost more than leaving it. A regulator requires a particular activity to sit in its own licensed entity. A partner would only come in at the level of one product line, and that line became a company. A country's law required a locally incorporated vehicle to hold a lease. Somebody's tax advice in 2011 pointed one way and nothing since has been worth the cost of reversing. Every one of those produces an extra box on a chart, and not one of them says anything about anybody's honesty.

Group structures arise from history, regulation, partnership and tax. Complexity is not evidence of concealment, and a reader who reasons from structure to motive has left analysis behind. How that error is built matters more than how it sounds. Five steps of real work sit in front of it, and their credibility flows into whatever sentence comes next. Establishing the statements, mapping the structure, locating the profit, taking out the outside share, listing the participations, and then writing that the arrangement looks designed to hide something takes standing earned by arithmetic and spends it on a claim the arithmetic never came near. The borrowed credibility is precisely why the guess lands harder than the same guess offered on its own with nothing behind it.

The first and the third collapse for related reasons. Suitability is a judgement about what a particular business requires. Reaching it needs the history, the licences, the bargains with partners and the plans, and none of those four is anywhere in a set of accounts. And an outside holding is simply a fact about who holds shares in a company. On Anjani Stationers, the 30 per cent held outside Chitra Binding Works produces a Rs 10,50,000 line in equity and a Rs 3,00,000 line in profit, and it states exactly that and nothing more. The outside holding does not say the parent was unable to buy the rest, did not want to, or is planning to. Each of those is a fact about a negotiation, and negotiations are not in the accounts.

Anjani Stationers Private Limited, Chitra Binding Works and Vaidehi Rao are invented.
Educational material. Not advice on any investment, tax, budget or market position.