Financials

Financials — what the numbers say

Figures converted from INR at historical FX rates — see data/company.json.fx_rates. Ratios, margins, multiples and growth rates are unitless and unchanged.

Harsha is a small-cap precision-engineering exporter (FY26 consolidated revenue $173M, market cap $383M) that finally re-accelerated after three flat years: FY26 revenue grew 16%, EBITDA grew 51% and net income grew 74% YoY, with Q4 alone clocking 27% top-line and 15.4% operating margin — the best print in 13 quarters. The balance sheet is a quiet strength (net cash since the September-2022 IPO, $50M of treasury investments versus $40M of borrowings), but earnings quality is mixed: cash conversion is volatile because the business runs on a ~170-day cash cycle and is mid-cycle through a capex push (Advantek phase-2, China expansion, $12M capex in FY26 with another $14-15M guided for FY27). The stock trades at 23.9× trailing P/E and ~12.9× EV/EBITDA — roughly in line with NRB Bearings (the closest pure-play peer) and at a sharp discount to Schaeffler/Timken/SKF India bearings OEMs at 50-70× P/E. The financial metric that matters most right now is sustained Engineering-segment EBITDA margin: management has guided 100-200 bps of improvement over 2-3 years off the FY26 base, and without it the multiple has no clear path to re-rate.

Headline KPIs

1. Financials in one page

Revenue compounded at 6.8% over the six fiscal years FY20 → FY26 in dollar terms — but the path was bimodal. FY20 to FY22 grew revenue 49% ($117M → $174M, helped by the bearings post-Covid recovery). FY22 to FY25 then plateaued ($174M → $165M — actually down 5% in dollars while modestly up in rupees) as Europe weakness and the loss-making Romania subsidiary capped earnings. FY26 broke that range: $173M revenue (+16% YoY in rupees), $29.7M EBITDA (+51%), $16.5M net income (+74%). Operating margin recovered from a 12.4% trough in FY25 to 14.7% in FY26 — and management says it can rise another 100-200 bps over two to three years as Advantek and China utilization ramp. The balance sheet is small-cap-conservative: $40M of borrowings versus $50M of cash and investments leaves the company in a net cash position, supporting a heavy capex programme without raising leverage. Free cash flow is the soft spot — negative in FY22, FY25 and FY26 because of working-capital absorption and growth capex. The valuation is the second soft spot — 23.9x P/E for a 14% ROCE / 12% ROE business is fair, not cheap. The single metric to watch is Engineering-segment operating margin, currently ~16% at the segment level and the lever for any further re-rating.

2. Revenue, margins, and earnings power

Definitions, once.

• Revenue = consolidated top-line including Engineering (cages, bushings, stampings) and Solar (mounting structures).
• EBITDA = earnings before interest, tax, depreciation and amortization — the cleanest pre-capital-structure measure of operating cash generation.
• Operating margin (here) = operating income / revenue. Net margin = net profit / revenue, after interest and tax.

Six-year revenue, EBITDA, net income

Revenue is the steady part of the story. EBITDA and net income are the volatile parts — and that volatility is the investment question. FY23 net income of $15.0M was the prior peak; FY24-FY25 reset to $13.3M and $10.4M as Europe slowed, Romania remained a drag and the standalone India business absorbed the post-IPO depreciation step-up. FY26's $16.5M broke through.

Margin structure — gross, operating, net

Read this chart as the central financial debate. FY23 was the cyclical peak (16.1% EBITDA margin). FY24-FY25 retraced to a 13-14% range as freight costs, Romania losses and product-mix dilution from the Solar entry compressed margins. FY26 marks a re-acceleration: 17.2% EBITDA margin — a new high — driven by a richer mix (large-size cages +14%, Japan customer +12%, bushings +25%), better export pricing, pass-through of brass costs and Advantek turning EBITDA-positive. Management is guiding another 100-200 bps over the next two to three years. If that holds, FY28 EBITDA margin would land at 18-19%, competitive with the Indian bearings majors.

Quarterly trajectory — the inflection is recent

The most important quarter on this chart is Q4 FY25 (operating margin 9.4%) — that low blew up the FY25 net income to $10.4M and triggered the share-price drawdown into FY26. Q4 FY26 is the mirror image: 27% revenue growth, 15.4% operating margin, $5.0M net income (versus a $-0.2M loss in Q4 FY25). The four-quarter sequence — 15.1%, 14.3%, 13.9%, 15.4% — is the strongest evidence that the FY24-FY25 trough was cyclical, not structural.

3. Cash flow and earnings quality

Define free cash flow once. FCF = operating cash flow minus capex. It is the cash the business produces after running and reinvesting in itself — what's left to repay debt, pay dividends, buy back shares, or make acquisitions. A multi-year pattern of net income consistently above FCF is a red flag: it means accounting profit isn't turning into bank-account cash.

Net income vs operating cash flow vs FCF

What the cash-flow gap tells you

Cash conversion — the honest answer

The implication for valuation: the trailing 23.9x P/E is not a 23.9x P/FCF. On normalized free cash flow (~60-70% conversion assumption post-capex cycle), Harsha trades closer to 35-40x current-year FCF. That gap is the principal reason this stock is not a screener-cheap stock and should not be analyzed purely on P/E.

4. Balance sheet and financial resilience

Borrowings, equity, and net debt — the IPO inflection is visible

Pre-IPO Harsha had $55M of borrowings against only $49M of equity — a 1.1x debt/equity ratio that would have made the FY24-FY25 margin compression genuinely dangerous. The September 2022 IPO raised roughly $60M of fresh equity, which was applied to retire borrowings (borrowings fell from $51M in FY22 to $22M in FY23). Since then borrowings have crept back up — partly because of working-capital lines for growth, partly because the Advantek + China capex is being funded with a debt-cash mix. Net debt remains negative because treasury investments ramped in parallel ($36M in FY25, $50M in FY26).

Liquidity, leverage, and interest cover

The leverage profile is comfortable: FY26 interest cover is 17.4× (EBITDA $29.7M over interest $1.7M), debt/equity is 0.27x and net debt is negative $11M. There is room — and stated intent in the Q4 transcript — to add another $14-15M of capex in FY27 without external equity. The balance sheet is the single best thing about this stock.

Working capital — the unglamorous risk

This is the most under-discussed risk in the file. Inventory days are stuck at ~165-170 — the realistic floor for a global precision-manufacturing supplier that holds 5-6 months of work-in-progress and finished cages for Tier-1 OEMs. Receivable days are 85 — concentrated in Schaeffler/SKF/Timken-class customers — and not easily compressed. The CFO confirmed on the Q4 FY26 call that working-capital cycle is ~130 days at the consolidated level, with the FY26 capex-and-growth wave widening total working-capital days to 147 from 97 a year earlier. Every $10M of revenue growth needs ~$4.0-4.5M of working capital — which is why operating cash flow lagged net income by $9.3M in FY26.

5. Returns, reinvestment, and capital allocation

Return on capital employed — improving, not exceptional

ROCE peaked at 18% in FY22 (pre-IPO, when the equity base was much smaller) and bottomed at 12% in FY24-FY25 as the IPO equity sat partly idle in treasury and Romania bled. FY26's 14.3% ROCE is mid-cycle — and it is the right metric to track because ROE understates returns while the company is sitting on $50M of treasury investments earning a money-market yield. Adjusted for excess cash, operating ROCE is closer to 18-20%.

Capital allocation since IPO — where the $55M+ went

The post-IPO capital story is direct: roughly $55M of cumulative capex (Romania upgrade, China expansion announcement at $9.94M, Advantek phase-1 and phase-2 footings, solar capacity) and a modest dividend that has crept from 7% payout in FY23 to ~10% in recent years. There have been no buybacks and no acquisitions. Share count has remained essentially fixed at ~91M shares since IPO; per-share EPS growth is therefore an honest reflection of earnings growth (FY23 $0.16 → FY26 $0.18 = +13% cumulative over three years in dollar terms; +26% in rupees).

EPS and dividend payout

The verdict on capital allocation is benign rather than enthusiastic. Management is reinvesting most of the free capital in capacity — China expansion, Advantek, large-size cages, bushings — and keeping dividends at a low single-digit payout. There is no buyback culture. That's fine if the capex earns above-cost-of-capital returns. The honest read is that we won't know whether Advantek phase-2 and the China expansion are accretive until FY28; until then, this is faith-based capital allocation backed by a long operating history and a 50-60% domestic market share.

6. Segment and unit economics

Disaggregated segment financial JSON was not available in the data pack; segment colour is reconstructed from the Q4 FY26 management commentary. Harsha reports two segments — Engineering (bearing cages, stampings, bushings) and Solar (mounting structures for utility-scale solar). The chart below uses management-disclosed full-year figures.

Three things matter in this table.

One. Bearing cages still drive ~73% of revenue and most of the profit. India Cage business grew about 11% in FY26; large-size cages grew 14% to $5.2M and the Japan customer grew 12% to $7.8M. These are small numbers in absolute terms — large-size cages are 3% of revenue today — but they are the highest-margin growth vectors and the main reason management is guiding 100-200 bps of margin expansion.

Two. Bushings is the second growth vector — $13.5M in FY26 (+25% YoY) with guidance of 25-30% growth in FY27. Cross-checked against Menon Bearings (the only listed bushing pure-play in our peer set), bushings are a ~20% EBITDA margin business at scale.

Three. Solar is a low-margin add-on. $19.5M revenue, $1.1M PAT (~5.6% net margin). It diversifies the order book and absorbs surplus engineering capacity, but it dilutes the Engineering EBITDA margin when consolidated. The FY26 adjusted Engineering-segment EBITDA was $28.8M on $153.9M of segment revenue — an 18.7% Engineering-only EBITDA margin, materially better than the 17.2% consolidated figure.

7. Valuation and market expectations

Current price implies a benign re-rating, not a deep cycle

The right valuation lens for a precision-engineering exporter with 14% ROCE, 12% ROE, a meaningful capex cycle ahead and choppy cash conversion is EV/EBITDA cross-checked against P/E and P/B — not P/FCF (because FCF is depressed by transient capex) and not EV/Sales (because the business is profitable).

History — limited window, but informative

The stock IPO'd in September 2022 at $3.40 issue price; the historical multiple range is therefore only ~3.5 years. The trailing P/E has oscillated between ~20x (Mar 2023, Mar 2025) and ~50x (Jul 2024 peak when EPS troughed during the FY24 margin compression). At 23.9x today, the multiple is at the lower end of the post-IPO band — and it sits there because the market is pricing FY27 EPS growth, not stretching for FY28.

Sell-side consensus

Consensus average target ~$4.90, +17% upside on the May 21 close of $4.20. Prabhudas Lilladher (the most cautious of the major brokers covering the name) values the company at 20x March-2028E EPS — the simplest summary of the bull case. The bear case is that 20x is already a 1.4x PEG on a 14% long-run earnings growth assumption.

Bear / base / bull, sketched on FY28 earnings

The asymmetry is mildly positive: ~30% base-case upside on a single year of multiple stability, ~80% bull-case upside on the full margin guidance landing. The bear scenario does not require a collapse — it just requires the FY26 margin print to be a top, with growth slowing to single digits.

8. Peer financial comparison

The peer set is the only available comparable group: four listed Indian bearings names. Three of them (Schaeffler, SKF, Timken) are also Harsha's customers — they make the bearings, Harsha makes the cages inside the bearings — so the relationship is customer-competitor, not pure peer. NRB Bearings and Menon Bearings are pure plays and closer comparables on margin / capital structure.

Three peer-comparison observations matter.

One. Harsha trades at 23.9x P/E and 12.9x EV/EBITDA — almost identical to NRB Bearings (23.6x P/E, 13.7x EV/EBITDA) and slightly above Menon Bearings (21.5x P/E). The market is grouping Harsha with the small-cap Indian pure plays, not with Schaeffler/Timken/SKF India.

Two. The valuation gap to Schaeffler (51.6x P/E), Timken (70.7x) and SKF India (53.4x) is enormous — but it's explained by ROCE: Schaeffler runs 27.9% ROCE, Timken 18.3%, NRB 18.6%, Harsha 14.3%, SKF 13.3%. The multiple ranking lines up almost perfectly with ROCE ranking (Schaeffler > Timken ~ NRB > Harsha > SKF). Harsha would need ROCE in the 18-20% range — achievable if FY26 margin expansion holds and Advantek/China scale — to justify a re-rating closer to Timken's multiple.

Three. Harsha's EBITDA margin (17.2%) is competitive with the listed Indian OEMs (NRB 19.4%, Schaeffler 20.7%, SKF 21.0%, Timken 18.7%) — but the OEMs have higher ROCE because their asset turnover is faster. Cage manufacturing is more capital-intensive than finished bearings; this is structural. The fix for Harsha's relative multiple is operating leverage on existing assets (Advantek, China utilization), not margin parity.

9. What to watch in the financials

Closing read

The financials confirm three things. First, the IPO did its job: Harsha is a small-cap with bank-account flexibility, a comfortable balance sheet and the ability to fund a capex cycle internally. Second, FY26 is a credible inflection — 16% revenue growth, 51% EBITDA growth, 74% net income growth, 15.4% Q4 operating margin — and not just a base-effect bounce. Third, the multiple is reasonable rather than cheap: 23.9x P/E and 12.9x EV/EBITDA price in continued margin expansion and earnings compounding at low-to-mid teens.

The financials contradict one comforting narrative — that this is a "cash compounder." Over the seven-year window FY20-FY26, cumulative free cash flow was only 32% of cumulative net income. Until the capex cycle normalizes (FY28 at the earliest), the right way to underwrite this stock is on the quality of earnings growth rather than the absolute level of free cash flow.

The first financial metric to watch is the Engineering-segment EBITDA margin in the FY27 quarterly prints. If it holds above 19% (versus 18.7% FY26 adjusted) the bull case stays alive and the multiple has a path to 26-28×. If it slips back below 17%, the FY25 multiple range (~20×, implying ~$3.50) becomes the reference and a fresh catalyst is needed to lift it.