Synergy or Cost Saving: The One Question That Sorts Them

A cost saving is any reduction in what a business spends. Synergy is the narrower thing: the part of a saving that exists only because both businesses have come under one ownership. One question sorts them. Could this change have been made without the purchase? If yes, the saving is real and welcome, and it is not a reason to have paid anything extra.

The distinction reads as a small one, and a great deal of money turns on it. A buyer hands over a price today for a business it will hold for years, and the price is argued for using a list of things that will get better. Some of those things needed the purchase to happen. Some of them were sitting on the buyer's own desk the whole time. Both kinds improve profit, both kinds are worth doing, and only one of them is a reason to pay a rupee more than the business was already worth. Everything below is about telling them apart, and about what it costs when nobody bothers.

What exactly is a cost saving?

A cost saving is money that used to leave the business and now does not. There is nothing more to it than that, and the plainness is the point. A chaiwala outside a college gate finds a milk supplier two streets over who sells at four rupees a litre less. He buys the same milk, makes the same tea, sells at the same price, and keeps more at the end of the day. The four rupees a litre is a cost saving in its entire form. Nobody bought anybody. No transaction was needed. He simply spent less.

A cost saving is straightforwardly good for whoever makes it. The point is worth making early because the sorting below keeps pushing cost savings out of one column and into another, and that is easy to hear as a judgement about their worth. It is not. The chaiwala who saves four rupees a litre is four rupees a litre better off, and asked whether his saving was a synergy he would rightly wonder what the question was for.

Cost savings arrive constantly and from every direction. A supplier is renegotiated. A contract that renewed automatically is finally read. Two people doing overlapping work become one. A lease is not renewed because half the floor was empty. A licence nobody uses is cancelled. Harivansh Packaging Limited, a maker of rigid and flexible packaging listed on both Indian exchanges, will find several of these in any given year without a transaction anywhere in sight, because every business of that size is carrying spending it would not choose again today.

Notice what none of that required. Not one of those savings required a purchase, a board approval for a purchase, a lender, or a price. Each one required somebody to look, decide and act. Looking, deciding and acting is the ordinary machinery of running a business, and it produces the great majority of the savings any company ever makes.

What is synergy, once a cost saving has been defined?

Synergy is a subset of cost savings, and a much smaller one than most transaction papers suggest. Synergy is the part of a saving that could not have been obtained by either business acting alone. Not the part that was harder alone, not the part nobody had got to yet, and not the part that is larger now. The part that was genuinely unavailable until both businesses came under one ownership.

The definition is deliberately narrow, and most items presented as synergy do not survive it. Narrowness is precisely why the definition is worth having. A test that everything passes is not a test. If the word covered every saving that appeared after completion, it would be a synonym for cost saving with a more impressive sound, and it would carry no information at all. Its whole usefulness comes from how much it excludes.

Think about what the combination actually changes. Two businesses that were separate are now one. The combined business files once instead of twice. A shared supplier now sees one contract instead of two. One of two facilities that served the same route can close. One function can run where there were two. Every one of those changes exists because the boundary between the two companies stopped existing. No amount of effort inside either company on its own could have produced them. A company cannot reach across a boundary that is still there.

Everything else is a cost saving that happened to be found during a busy period. Sundarban Polymers Private Limited, unlisted and making films for flexible packaging, has spending it could reduce on its own, exactly as Harivansh Packaging Limited does. Being bought does not create those opportunities. The purchase creates a moment when somebody finally looks at them. A moment of attention is a different thing from an opportunity, and it is worth nothing extra in a price.

What is the single question that sorts them?

Could this change have been made without the purchase? Eight words are the whole test. The question applies to every item ever proposed, and it is the only thing standing between a disciplined price and a wishful one.

The question is answered about the change, not about anybody's intentions. This is the part people get wrong, and they get it wrong in a way that always moves in the same direction. Somebody will say that yes, technically the target could have renegotiated that contract alone, but honestly they never would have. The claim feels like an argument and is not one. Whether a particular set of people would eventually have got round to a particular task is unknowable. Nobody can settle it, nobody can be held to it, and a price built on it rests on a claim that can never be checked.

Take it out of the transaction room. A household in Pune buys the flat next door, planning to knock through and have the space they have wanted for years. In arguing themselves into the price, somebody says that once they have moved they will stop ordering food four nights a week and save real money every month. The saving is real and the decision is a good one, and neither has anything whatever to do with the flat. The household could have stopped ordering food last Tuesday. If the saving on food is part of why the flat seemed affordable, they have paid the seller for a decision that was always theirs to make.

The test also has a useful property: it produces the same answer regardless of who is asking. Two people who disagree about whether a target's management is any good will still agree on whether a particular supplier contract could have been renegotiated by one business alone. Volumes and terms settle that one, not character. Anchoring the sorting to a factual question is what keeps it from becoming a negotiation about optimism.

Why does the difference change what a buyer should pay?

Take the logic a step at a time. Each step is obvious on its own, and the conclusion is not.

A buyer pays a price for a business. The price is the value of the business the buyer receives. Whatever the buyer could already have achieved without buying, it held before the transaction started. The purchase delivered none of it. So a price argued for on the strength of such an item has been argued for using something the buyer already held. The buyer has paid the sellers for an asset that was never theirs to sell.

The conclusion is severe enough to be worth taking slowly. Nobody has been defrauded. The saving is genuine and it will arrive. The item is simply on the wrong side of the table: it belonged to the buyer before anybody signed anything, and it has been used as a reason to send money in the other direction.

The same mechanism is how a purchase price grows during a process while nothing whatever about the business being bought has changed. Watch how it happens. The first cut of the case supports a certain price. The price is not enough to win. Another round of work goes into finding savings, and it finds them. Any business of any size has savings sitting in it. The new items go into the same column as the old ones, the column total rises, and the price rises with it. From the outside this looks like better analysis. On some of those items, though, the buyer is simply paying for its own opportunities.

What that costs in the Harivansh Packaging purchase

Put figures on it. Harivansh Packaging Limited buys the whole of Sundarban Polymers Private Limited, and the enterprise valueThe price of the whole business including the borrowings it carries, before anything is taken off for what it owed. of the target is set at Rs 1,320 crore, or 10.0 times its earnings before interest, tax, depreciation and amortisationA profit line struck before interest, tax and the depreciation charge, so trading performance can be compared without funding and asset age getting in the way., or EBITDA, of Rs 132 crore. Take off Sundarban Polymers Private Limited's net debtWhat a business has borrowed, less the cash it is sitting on. It is the borrowing that would survive if the cash were used to pay some of it down. of Rs 180 crore and the equity valueWhat the sellers walk away with for their shares, once the borrowings sitting inside the business have been dealt with inside the price. paid to the sellers is Rs 1,140 crore. The equity value is the amount that leaves Harivansh Packaging Limited and reaches human beings. Behind it sit two sources: Rs 140 crore drawn from cash the acquirer was already holding, and a further Rs 1,000 crore borrowed fresh at a contracted 9.0 per cent.

The cost of that money can now be weighed against what it bought. Sundarban Polymers Private Limited contributes profit after tax of Rs 61 crore. The Rs 61 crore has been rounded. Working the chain exactly produces Rs 61.35 crore, and Rs 61 crore is the value carried throughout. Rs 61 crore set over the Rs 1,140 crore handed to the sellers gives an earnings yield of 5.35 per cent. A year of the borrowing costs Rs 90 crore before tax, and once a 25.0 per cent effective rate has taken its share the after tax bill is Rs 67.5 crore, and putting that over the identical Rs 1,140 crore gives 5.92 per cent. One base serves both sides. A cost measured against only the Rs 1,000 crore borrowed cannot be set beside a yield measured against the whole price and be expected to tie.

Set 5.35 against 5.92 and the shortfall is 0.57 of a percentage point. Applied to Rs 1,140 crore, that shortfall is Rs 6.50 crore less profit after tax than the money cost, and spread over 18.00 crore shares it works out at Rs 0.36 off each share. Harivansh Packaging Limited's earnings per shareOne year of profit after tax spread evenly over every share in issue, so each share can be said to stand behind a stated amount. is therefore reported at Rs 12.14/- where it used to be reported at Rs 12.50/-, dilution of 2.9 per cent. Adding the parts reaches the same place: Rs 225 crore of the acquirer's own profit after tax, Rs 61 crore contributed by what it bought, Rs 67.5 crore of after tax interest taken away, leaving Rs 218.5 crore.

Closing that gap needs Rs 8.67 crore of extra EBITDA, and the sorting question decides which items are allowed to count toward it. The arithmetic runs backwards from the target: Rs 12.50/- across 18.00 crore shares needs Rs 225 crore of profit after tax, the shortfall is Rs 6.50 crore, and at a 25.0 per cent effective rate the operating line has to produce Rs 8.67 crore for Rs 6.50 crore to survive to the bottom. Measured against Rs 609 crore of combined EBITDA, an uplift of Rs 8.67 crore is 1.42 per cent. An uplift that small is genuinely achievable, and achievable in two quite different ways, of which only one means anything.

Where does the confusion actually come from?

The confusion is structural rather than dishonest, and treating it as a character problem is why it never gets fixed. Picture the people producing the number. A small team, several weeks into a process, working to a Friday board paper. The team has a list of identified savings gathered from operations, procurement and finance. The template they were given has a column headed synergies, a column for the annual amount and a column for the year it lands. There is no column asking whether the purchase was necessary. There almost never is.

So every item goes in the one column that exists. Nobody decided to overstate anything. The form had one place to put a saving and the savings went there. By Monday the total is in a board paper, by the following month it is in a price, and nobody has asked what would have split the list in two. Nothing on the form asked it.

The problem gets worse in a quiet way. Two items that look identical in a spreadsheet can be completely different things. A supplier renegotiation worth some amount a year, landing in year one, sits on one row. Another supplier renegotiation worth a similar amount, landing in year one, sits on the row below. One of them needed the combined volume and the other did not, and nothing visible on either row shows which is which. The spreadsheet is not hiding the difference. The spreadsheet simply has no way to express it.

The fix is a column, not a stronger conscience. Add a field that records, for every line, the answer to the sorting question and one sentence of reason. One column is the whole intervention. The column costs a few minutes per item, it survives staff changes, and it turns an unanswerable argument about who was being optimistic into a row that either has an answer written on it or does not. Anything that relies on people being careful under deadline pressure will fail eventually; a required field fails only if somebody deletes it.

What happens to an item that sits on the line?

Some items refuse to answer cleanly, and pretending otherwise is how a register loses its credibility. Two shapes come up again and again.

The first is scale. A saving was technically available to one business alone and was not worth the effort at that size. A system change costing a fixed amount to implement pays back over the volume it touches, and at one business's volume the payback was too long to justify. At combined volume it is comfortable. Was that saving available alone? Technically yes. Was it a real option? Not really.

Two households marrying provide the everyday version. Each set of relatives could have negotiated with a caterer on their own. Neither had the guest count to be interesting to the caterer who does the good work at a serious discount. Held together, the combined function is worth a proper conversation. Nobody was prevented from calling the caterer before; the call simply would not have gone anywhere.

The second shape is access rather than permission. Sometimes the combination does not make a saving legal or possible so much as it makes it visible. A buyer that sees another business's supplier terms for the first time may discover it has been paying more for the identical input. The knowledge came from the transaction. The renegotiation, though, was always available. The honest description is that the purchase supplied the information and not the opportunity, and that is a weaker claim than a synergy but a stronger one than nothing.

The honest treatment is to record the judgement and the reason on the line, rather than to force a binary the case does not support. Which way the call went and why belongs in one sentence on the row itself. Somebody reading it in eighteen months can then disagree with the reasoning. Being disagreed with is exactly why the reason is written down. The alternative is a register where the borderline items were quietly resolved in whichever direction suited the price, leaving nothing on the record to argue with.

A register that records its borderline calls is more trustworthy than one that has none. A list of forty items where every single one is a clean yes or a clean no is not evidence of a rigorous process. A clean list is evidence that the awkward ones were pushed into a box. Judgement showing its working is the strongest signal a document of this kind can carry.

What does a reader look for in a report that claims savings?

A synergy case is read far more often than it is built, in a results presentation, a lender update or a quarterly programme report. Three things are worth checking, and all three are absences rather than statements.

First, is the saving expressed against a baselineThe agreed starting level a change is measured against. Without one, a reported improvement has nothing behind it to be an improvement on.? A saving is a difference between two numbers, and a report that supplies only one of them has supplied an assertion. Against what the cost was last year, or against what it was budgeted to be, or against what it would have been had volumes not fallen: those are three different baselines and they produce three different savings from identical facts.

Second, does the cost to achieveThe money spent to obtain a saving: the redundancy payment, the system change, the site closure. It is usually spent once while the saving repeats, which is why the two are reported separately. appear anywhere? Savings are rarely free. Redundancy costs money, closing a site costs money, moving two systems onto one costs a great deal of money. A programme reporting savings gross of what it spent obtaining them has told half the story, and it is the flattering half.

Third, does anything in the document distinguish the two categories at all? A report using the word synergy for every saving it has found has told the reader that no test was applied. That is not a small inference and it is a reliable one. If the sorting had been done, the document would show it. A team that has done that work is proud of it and says so. Silence on the split is evidence about the process rather than about the savings.

One more habit worth having. Savings are often quoted on a run rateA figure scaled up to a full year from a shorter stretch, so that a change made in month nine is quoted as though it had been in place for all twelve. basis. Quoting a run rate is legitimate and also the easiest place for a number to look larger than the year it sits in. A run rate saving and the saving that actually lands inside one reported year are different quantities, and a document that moves between them without saying so is worth reading twice.

How do these items sort in the Harivansh Packaging purchase?

Take six ordinary items of the kind any transaction of this shape produces, and run each one through the question. Before the first item, though, look at what the two businesses have in common. The similarity closes off an entire line of argument before anybody opens it.

On revenue of Rs 3,180 crore, Harivansh Packaging Limited earns EBITDA of Rs 477 crore, so one divided by the other gives a margin of 15.0 per cent. Do the identical sum for Sundarban Polymers Private Limited, whose Rs 132 crore of EBITDA sits on revenue of Rs 880 crore, and the margin comes back at 15.0 per cent again. Neither business is the demonstrably better operator, so no item on the list can be justified by claiming that one side simply runs its costs better than the other. That argument is extremely common and it is unavailable here. Put the two together and revenue reaches Rs 4,060 crore against EBITDA of Rs 609 crore, still 15.0 per cent. There was never a gap between them for the combination to close.

Now the six items, with the reasoning shown rather than the answer asserted. None of the six carries a rupee amount. The sorting turns on whether the purchase was needed and not on how large the item is, so a figure would decide nothing and would only make the list look more settled than it is.

The fourth item is the commonest borderline in any transaction of this kind, and it is the one most often resolved in whichever direction suits the price. Notice that resolving it either way is a decision, and only one of the two ways leaves a trace. Pushing it into the synergy column raises the supportable price and writes nothing down. Recording it as a call with a reason leaves something a later reader can argue with. The second is harder in the meeting and far stronger in eighteen months.

Now put the sorting and the price side by side

Suppose the case for the Rs 1,140 crore leaned on all six items rather than on three. The two failing items still deliver. The savings arrive on schedule, the profit line improves, and every progress report shows the programme in good health. And both of those savings were available to Harivansh Packaging Limited without spending Rs 1,140 crore.

The business ends up better off and the purchase still did not pay for itself, and both statements are true at the same time. Both statements are worth carrying away together. Delivery and justification are separate questions, and a programme can answer the first one perfectly while the second one was never asked. Nothing in the reporting afterwards can tell the two apart. The reporting only measures delivery.

One last stretch of arithmetic, on how a saving gets counted

Take an item reported as a Rs 3 crore saving, where obtaining it cost Rs 2 crore. The business keeps Rs 1 crore. Two thirds of the headline went on getting it, and a report showing the Rs 3 crore has described the saving without describing the price of the saving.

Now stack the two errors. An item that fails the sorting question, counted gross of what it cost to obtain, contributes Rs 1 crore to a business that would have had it anyway, while being presented as Rs 3 crore of value the purchase created. The overstatement is not one mistake sitting on top of another; it is two different mistakes multiplying, and each one is invisible in a report that has neither a starting level nor a cost line.

How does someone outside the transaction use this?

Four kinds of reader come to a savings number with four different jobs, and the sorting question serves all of them.

A lender sizing a facility against a purchase is asked to believe a repayment profile that depends partly on promised savings. The lender's interest is narrow and sensible: how confident can it be that the money arrives. Items that pass the sorting question have a mechanical reason to arrive. The mechanism is the combination itself, and the combination has already happened. Items that fail depend on somebody doing work they could have done at any point in the past several years and did not. The second category is not worthless to a lender, and it is a different kind of promise, and knowing which is which changes how much comfort a schedule deserves.

An equity analyst reading a results presentation from an acquirer has a narrower question still: is the improvement in this year's profit something that will repeat, and did the purchase cause it. A saving that would have happened anyway is going to show up in the numbers regardless, and attributing it to the transaction makes the transaction look like it worked. The analyst's protection is the sorting question applied from the outside, and it is applied by asking what the acquirer could have done alone rather than by trusting the label on the chart.

An investor holding shares in an acquirer meets the same mistake in the form that costs real money. The price was paid out of value that already belonged to those shares. If part of that price rested on savings the business already had access to, the transaction transferred value out to the sellers and delivered improvement that was coming anyway. The transfer will not look like a loss in any report. The savings are genuine and the profit does rise. The loss shows up instead in a return on the money spent that never quite improves, and slow and quiet is the only way a mistake of that kind ever becomes visible.

And a household faces exactly the same question at a much smaller scale, whenever it justifies a purchase using a saving it could have made without buying anything. The rule is identical at every size: a change that was always available is not a reason to hand money to somebody else. Buy the thing if the thing is worth its price. Make the saving because the saving is worth making. Keep the two decisions apart, and neither one can be used to excuse the other.

Harivansh Packaging Limited, Sundarban Polymers Private Limited, Devyani Kulkarni and Ashwin Rege are invented.
Educational material. Not advice on any investment, tax, budget or market position.