Capacity was capitalised and financed faster than it was sold

Across 7 calls between 2021-07-30 and 2025-01-17, commissioned capacity moved from 72 MW to 120 MW. The flat stretches are the finding, so nothing here softens them.

This is the same estate the rest of the page reads out of the prospectus, described in different words. On the call the figure is what has been commissioned. In the filed document the widest number is built capacity, and it is 188.04 MW. The promises marked here carry the date they were made rather than a verdict, because that is what the record supports.

Across the four filed periods the estate absorbed 2,310 crore against 1,233 crore of cash from operations, so the building ran at 1.87 times what the business produced and the difference, 1,077 crore, came from somewhere else.

The single covered period is not a change of habit. Capital expenditure fell 55 per cent in FY2025 while operating cash rose 159 per cent, and the quarter that follows it is back to spending more than it earns. Capex leads commissioning by years, so a year of low spending is not a year of low building. It is a gap in the trail, and the offer is what fills it.

On the first measure this issuer is the opposite of a cash conversion problem. Profit after tax turns into between 2.26 and 7.39 times as much operating cash across the filed periods, which is what a business with heavy depreciation and customers on contract looks like.

The second measure is the one to read carefully, and it is why both are shown. Its numerator adds investing cash back in, so on an operator part way through building an estate it moves with the size of the build rather than with the quality of the earnings. It sits at 22.8 per cent in FY2023, standalone, near the threshold, and falls away in the year capital spending fell. Nothing about the earnings changed between those readings. The spending did.

The basis beside each period is the document's own. All four columns sit under a heading that reads restated consolidated, and the column header printed above them says consolidated for the two most recent and standalone for the two older ones. The series still holds, because the associate contributed 0 in the standalone years, so a consolidated statement for those years would have been the same statement. Where it does register is the stub, whose share of the associate's loss is 8.4 per cent of the profit reported for that quarter.

• 2 per cent of net worth348.17
• 2 per cent of turnover285.67
• 5 per cent of average absolute profit after tax, last three years52.72binds

The policy takes the lowest of the three, so the bar is 52.72 million rupees, 0.37 per cent of a year's revenue, and the next test up sits 5.4 times higher. Taking the lowest is the inclusive choice and it is worth saying so: a net worth test alone would have set the bar at 348. It also means the bar moves with earnings rather than with the size of the company, so a more profitable year raises it.

Against that bar, the largest matter carrying a number is 1,175.83 million, 22 times the threshold. The more interesting entries carry no number at all.

22 proceedings reached this issuer through a public database rather than through service, in its own words on both counts. Twenty of them carry no amount, because the grounds and the quantum are not available to a party nobody has served. So the largest number in the legal section is the largest number that could be written down, which is a different thing from the largest exposure in it.

Across these years the subsidiary's revenue grows 40 per cent and the gap moves by 0.14 million. A difference that holds its size while the business it sits inside grows by two fifths is a fixed item on one side of a boundary, not a measurement drifting from another.

It also survives the change of reporting basis. Two of these years are the subsidiary standalone and one is consolidated, and the gap does not notice, which is the second thing suggesting it belongs to the definition of the segment rather than to the perimeter of the company.

Neither document cites the other and neither reconciles the two, so what the item is cannot be named from either of them. What can be said is that the parent's segment and the subsidiary's accounts are not the same measurement, and anyone treating one as a stand in for the other is out by a fixed amount in every year they overlap.

The formula sits on printed page 261, inside the industry report the issuer commissioned. The figures sit in the business section, as the issuer's own achievement. The document never joins them, and joined they hold: every period that can be rebuilt lands on the published number to the second decimal.

It only holds on one reading. Capital employed is defined as net worth plus total borrowings less cash, and the definition does not say whether lease liabilities are borrowings. Counted, the rebuild is exact. Left out, as the words alone would have it, every period comes out 0.49 points too high. The stricter reading is the one the issuer used, which flatters it less, and the document does not say so anywhere.

FY2023 cannot be rebuilt at all. An average needs the capital employed of the year before it, and the balance sheet carries four columns. That period is also the highest of the four at 9.88 per cent, and it is half of what the claim to beating global peers rests on.

This company already appears twice in this analysis, once as the reason four columns of accounts can be read as one series. The statements are titled consolidated throughout and two of the four periods are standalone, and what keeps them comparable is that the associate contributed exactly nothing to profit in those two years. That is true and it is asserted as a build guard. Read alone it invites the conclusion that the associate does not matter.

The related party note says what else it is. At 2025-06-30 the issuer was owed 1,700 on loan, held 1,490 of its preference shares and had 96 out as a security deposit, and had guaranteed 1,387.5 more off the balance sheet. What that capital returned in the same period was -11.41, a loss. Immaterial to the income statement and 18.7 per cent of net worth are not in conflict. They are two different questions, and the document answers them in two sections that never refer to each other.