Board Independence vs Management Independence

Board independence describes the directors: how many of them are free of employment, ownership or commercial ties to the company and to the group that controls it. Management independence describes the executives: whether operating and reporting judgements are reached by them rather than settled for them. The first is counted off a disclosure. The second appears in no filing anywhere and must be inferred sideways, so the two are never one number.

Board independence and management independence opened this sequence side by side, and everything since has been building the equipment to finally separate them. The equipment is now in place: what has to reach the market and what a disclaimer is doing instead; an insider trade worked down to two thousandths of one per cent before a word is said about it; a promoter stake moving by 1.3 points without the record naming a single reason for the move. The reading method covered separately supplies three tests an observation has to survive, and the finding that better disclosure widens a range rather than shifting it. All of that is assumed here.

One job is left, and it is the job the whole sequence has been walking towards. Two phrases that sound like variations on a theme are not variations on a theme. The two phrases are about different people, they exist for different purposes, and they are evidenced in ways so unalike that only one of them produces a number at all. Almost every mistaken governance sentence written about a listed company comes from quietly using the countable one as a measurement of the uncountable one.

What exactly is board independence?

Board independence is a condition attaching to one director at a time. A director is independent when no tie binds them either to the issuer itself or to whoever controls it: no employment there, no ownership stake worth speaking of, no supply contract, no consultancy fee, no arrangement that would make disagreeing at a meeting personally expensive. The condition is tested person by person, and the test is about relationships that can be written down.

Because it attaches to individuals, it aggregates. Testing all nine directors and ticking the ones that pass produces a number for the board as a whole. Aggregation is why board independence has a percentage attached to it in every filing that reports one, and why management independence never does. Board independence is countable because independence is assessed against each director separately and directors are whole, separate, listable things.

Two things sit outside this guide, and both matter: which relationships disqualify a director, and how many independent directors a board must carry. Both are requirements rather than readings. The Securities and Exchange Board of India (SEBI) sets the listing side and the Ministry of Corporate Affairs sets the company law side, and both rulebooks move, so a written summary would be wrong the week it moved. The work here is reading a compositionThe make up of a board sorted into categories of director. A composition is published. How the board actually behaves in a meeting is not. that has already been classified, not classifying it afresh.

What exactly is management independence?

Management independence is a condition attaching to decisions rather than to people. The question is whether the executives running the business reach their operating and reporting judgements on the merits, or whether those judgements arrive already settled from the controlling group. The price the industrial line carries next quarter. Whether a slow moving stock of tinted base gets written down this year or next. Whether an awkward number goes into the disclosure with its full working attached. Every one of those is a judgement, and the question is where it actually got made.

Management independence cannot be counted. The unit is not a person, so there is no roll to tick. Management independence is a property of a process, nobody in any market counts processes, and this single difference generates every other difference in this guide. A judgement leaves no line in a register saying who reached it. The judgement leaves an outcome instead, and an outcome is consistent with several stories about how it was reached.

At kitchen scale the asymmetry comes out exactly. Two households can hold identical amounts in the same bank. In one, whoever is going to the market decides what the vegetables are worth that morning. In the other, the list and the prices are settled at home before anybody leaves. The bank statements at the end of the month can be indistinguishable. A statement records amounts and not who chose them, so nothing in it records which household it belongs to.

Who is each one independent of, and independent to do what?

Independence is never free floating. Independence is always independence from somebody, held so that something specific can be done. Get the second half wrong and the first half stops meaning anything.

A board is independent so that it can superviseTo oversee somebody else doing the work, on behalf of everybody with a stake in it rather than on behalf of whoever made the appointment. Oversight, not execution. the executives on behalf of every shareholder, including the small ones who will never attend anything. Supervision is the whole function. A director with no tie to the controlling group can ask an awkward question at a board meeting without it costing them their livelihood, and the awkward question is the product being bought.

Management is independent so that the operating choices and the reported numbers reflect what the business actually did, rather than what a particular owner would prefer to see. Execution is a different function entirely, and it sits one layer down. The board does not price the industrial line. The board asks why the industrial line is priced that way.

The two point in opposite directions along the same chain: the board looks downward to supervise, management looks outward to execute, and independence is doing a different job at each end. Take a one salary household again. The person who checks whether the money went where it was supposed to go and the person who does the actual buying need different kinds of freedom, and if the same person does both with nobody looking, the check is not a check. Neither freedom is a substitute for the other, and neither one implies the other.

How is each one evidenced, and why is only one of them counted?

Evidence is where the two part company for good.

Board independence is evidenced directly. A disclosure names the directors, sorts them into categories, and the categories are the very thing in question. Nothing is being inferred. The work is reading a classification somebody was required to publish and adding up the rows. The evidence and the question are the same shape, and the match is what makes a countA number produced by adding up disclosed items. A count is dependable about the items it added and completely silent about anything it did not add. possible at all.

Management independence is evidenced only sideways, by tracesAn observable consequence of something that was never itself disclosed. A trace is always consistent with more than one account of what produced it.. The judgement itself cannot be seen, so the analyst looks for things a judgement would have left behind. How much of the business runs through parties connected to the controlling group. Whether the senior executives are drawn from that group or hired from outside it. Whether the tone and completeness of disclosure shift when the controlling group has something at stake. Each of those is real, each is worth looking at, and not one of them is the thing itself.

A trace is weaker evidence than a count for one structural reason. A count settles what it counted. A trace always admits at least one other account of how it got there. Related partyA supplier, customer or lender connected to the company, or to the people in control of it. A connection is disclosed. Whether it changed any price is not. purchases exist in plenty of companies where every judgement is reached on the merits, because a connected supplier can simply be the supplier who is there. An executive drawn from the controlling group can be the person who knows the plant best. The trace narrows the question. The trace never closes the question, and nothing further obtainable from the public record will close it either.

Which brings the sharpest consequence in this guide. Governance figures that claim to summarise how a company is actually run are, taken apart, built out of countable things: board composition, attendance, committee membership, the presence of a policy document. Every one of those is on the board side of the line. Any number offered as a measurement of management independence is a count of board arrangements wearing a different label, and relabelling a count does not convert it into a measurement of something else.

Can a company have one without the other, in both directions?

Yes, in both directions, and this is not a theoretical caveat. Set the two out as a genuine possibility set and there are four cases, all of which occur.

Case one is the comfortable one: a high independent count sitting above executives who genuinely reach their own calls. Case two is a high independent count above executives who decide very little for themselves. The count describes who attends meetings and not what happens between them, so case two is entirely constructible. Case three runs the other way: a low count above executives who are independent in practice. In a company where the controlling group has deliberately stayed out of operations, case three is routine. Case four is neither.

All four cases occur, and the disclosed composition separates none of them. The two readings therefore cannot be collapsed into one. In ordinary terms: knowing how many people are on a school committee says nothing about whether the school buys its books from whoever the chair suggests. Both facts are real. Only one of them is written on the noticeboard.

The board of Sarvani Coatings Limited, worked for everything it gives

Sarvani Coatings Limited, an invented listed maker of decorative paints and industrial coatings, reports a board of nine directors, of whom four are independent, two come from the promoter groupThe controlling shareholders of a listed company, identified as such in its filings. A grouping disclosed by name and holding, not a judgement about anybody. and three are executive directorsDirectors who also hold a management post in the company. Sitting on both sides of the table, they are not independent of it by definition.. Work it properly. Work the counts, and never divide one printed percentage by another.

Two checks, and they behave differently. The seat check is clean: four plus two plus three is nine, and no seat is counted twice or left out. The percentage check is not clean, and pretending otherwise would be the sloppier choice. Four ninths, two ninths and three ninths each round downward to one decimal place, so the three printed shares add to 99.9 rather than 100.0. Unrounded they add to exactly 100. Quietly nudging one figure up to make a column add would teach the opposite lesson, so the parts are printed as they genuinely round and the shortfall is named. The two way split does add to a round hundred, since 44.4 rounds down and 55.6 rounds up, and that is arithmetic rather than luck.

The same composition, read for board independence, answers what it was built to answer. Four seats out of nine, or 44.4 per cent, are held by people with no disclosed tie to Sarvani Coatings or to its promoter group. The count answers who sits in the room, and answers it completely. The count is silent on what gets said there, and silent by construction rather than by omission. Whether 44.4 per cent satisfies any requirement is a matter for SEBI and for company law.

The identical composition, read for management independence, fails to deliver. The composition reports that three of the nine directors also hold management posts and that two come from the promoter group. The composition does not report whether the executives set the pricing on the industrial line, whether they choose what goes into the disclosure and in how much detail, or whether either of those is settled somewhere before it reaches them. No disclosure covers any of it. The composition is not an incomplete answer to that question. The composition is not an answer to it at all.

The traces available in the invented record are thin, and presenting them as thin is the honest move. Purchases of Rs 18 crore ran through an entity connected to the promoter group in year three, against a cost of materials of Rs 13,04,00,00,000/-, or Rs 1,304 crore. Work it in whole rupees and that is 1.38 per cent, leaving Rs 1,286 crore, or 98.62 per cent, going elsewhere. Sized against year three revenue of Rs 2,415 crore instead it is 0.75 per cent, and the two figures answer different questions, so the denominator has to be stated every time. The promoter group holds 51.1 per cent of the shares after the sale covered under promoter holding. A holding of 51.1 per cent is a majority of the shares in issue, and a majority is a statement about arithmetic, not about conduct.

One contrast is easy to draw badly, so draw it carefully. The promoter group holds 51.1 per cent of the shares and occupies 22.2 per cent of the seats. Shares and seats are two different denominators, and the gap between the two figures is not a finding about anything. The gap is what comes of dividing by two different things. Anybody who subtracts one from the other and reports the difference as a governance observation has produced a number with no referent.

Three notes on the arithmetic. Two of them would otherwise look like errors. First, every share of the board above was worked from the seat counts, four, two, three and nine, and never by dividing one printed percentage by another. Second, the three category shares print to 99.9 per cent rather than 100.0: four ninths, two ninths and three ninths each round downward at one decimal place, and unrounded they sum to exactly 100, so the parts are shown as they round and the shortfall is named rather than nudged away. The two way split prints to a round 100.0 because 44.4 rounds down and 55.6 rounds up. Third, the related party figure was worked in whole rupees, Rs 18,00,00,000/- over Rs 13,04,00,00,000/-, giving 1.38 per cent of the year three cost of materials and 0.75 per cent of year three revenue, and the denominator is stated wherever either appears. No such quantity is published, so the record carries no share of the board and no count of anything for management independence.

Why does merging the two produce a wrong reading?

Warnings are easy to nod at and mechanisms are not, so state the merge as a mechanism.

Suppose a reader treats the countable thing as a measurement of the uncountable thing. From that moment, every conclusion they draw about how decisions are actually made at the company rests on a number that was never about decisions. The conclusion may still be correct, in the way a stopped clock is correct twice, but it is no longer supported. Nothing in the file connects the evidence to the claim any more, and the disconnection is silent.

The reader ends up believing they hold evidence about the second thing when what they hold is a count of the first, and the error stays invisible precisely because a number was involved. This is what makes it different from an ordinary unsupported opinion. An unsupported opinion looks unsupported. A count of directors standing in for a judgement about process looks like measurement, reads like measurement, and gets filed like measurement. The assumption never announces itself, so it is never revisited.

What can legitimately be written about each?

About board independence a good deal can be written, and all of it is checkable. The count. The proportions, printed as they round with the rounding named. The fact that the classification came from a disclosure a stranger can open. And the boundary itself, stated rather than implied: composition is disclosed, conduct is not.

About management independence four lines can be written, and they have a fixed shape. First, what was looked for. Second, what was found. Third, how big it was, with its denominator stated. Fourth, the plain sentence that no measurement of the thing itself exists. Every sentence carries its trace, its size and the other account that fits it equally well, or it does not get written.

The honest paragraph about the board runs long. The honest paragraph about management runs to four lines. The shortness is a property of the evidence, not of the effort. That asymmetry is uncomfortable to hand over, because a short paragraph reads like a thin one. The short paragraph is not thin. Four lines is the correct length, and padding it is precisely how a researcher converts an absence of evidence into the appearance of a finding.