A software company sits on Rs 3,000 crore of idle cash. Compute the EPS effect of buying back shares at a premium and set out the case for a buyback, a special dividend and holding the cash.

A buyback of all the cash at Rs 165 lifts EPS from Rs 12.00 to about Rs 14.61, 22%; a special dividend cuts it to Rs 9.30 but hands each holder Rs 60 in cash; holding leaves it at Rs 12.00. The buyback works because the shares earn 7.27% at Rs 165 and the deposits earn 4.5% after tax. EPS accretion does not settle the question: the dividend returns the same cash without a premium, and holding is right only if the acquisitions are real and earn more than the cost of capital.

Because the cash earns less than the business it sits inside. Rs 3,000 crore at 6% earns Rs 180 crore before tax, Rs 135 crore after, so Rs 465 crore of Mridang's Rs 600 crore comes from operations and the rest from a bank account. The market pays 12.5x for the whole, but strip the cash out and it is paying only about 9.7x for the operating business, because nobody pays a multiple for a deposit. It is like a household with Rs 30 lakh in a savings account at 6% and a home loan at 9%: the money is safe, but every month it sits there it costs the family the difference.

Rs 3,000 crore at Rs 165 buys 18.18 crore shares, 36% of the company, leaving 31.82 crore. Net income falls to Rs 465 crore because the interest is gone. EPS is Rs 465 crore over 31.82 crore shares, Rs 14.61, up 22%. The reason is a yield comparison, not the share count: at Rs 165 each share bought carries Rs 12.00 of earnings, a 7.27% earnings yieldEarnings per share divided by the price paid for the share, the inverse of the P/E., while each rupee of cash gives up only 4.5% after tax. The buyback keeps lifting EPS at any price below Rs 266.67, where the earnings yield falls to 4.5%, which is a P/E of 22.2.

Because EPS accretion is arithmetic, and the board is asking about value. Under the dividend, each holder gets Rs 60 in cash and a share that now earns Rs 9.30; the total is the same Rs 3,000 crore, returned without paying anyone a premium. The buyback pays Rs 15 a share over the market price to those who sell, Rs 273 crore in all, and that comes out of the pockets of those who stay. It is a good trade for the remaining holders only if the shares are worth more than Rs 165, which is a view on the business, not a sum. There is also a practical point: 36% of the company cannot be bought in the market; it needs a tender offer, and holders decide whether to tender. The tax treatment of a dividend against a buyback in the holder's hands also differs, and changes with the budget, so tell the board to confirm the current rules before choosing.

Holding the cash is right in exactly one case: the acquisitions are real, priced, and earn more than Mridang's cost of capital, which is far above 4.5%. 'In the pipeline' is not that. A board that has sat on Rs 3,000 crore for years has already told the market it has nothing to buy, and the ex-cash multiple of about 9.7x is the market's reply.

A split, with a reason for each part. Return most of the cash, keeping perhaps a year of operating cash for a named deal with a deadline; if the deal is not signed within the year, return that too. Use a buyback only if the board believes the shares are cheap at Rs 165, and run it through a tender so every holder can choose; otherwise pay a special dividend, which treats every holder alike and pays no premium. Say plainly that the 22% EPS lift is the result of giving up a 4.5% deposit, something any company with cash can do, and that the market will not pay 12.5x for it twice.