AMRT Why Did Profit Keep Falling After the Growth Engine Recovered?

A Growth Engine That Came Back, and an Overhead Line That Didn't Hold

The last post on this company closed Q2 2020 - the first full lockdown quarter - with every one of Alfamart's three geographic segments losing both revenue growth and margin simultaneously, including Outside Java, the region every post since FY2018 has named as this company's real growth engine. This is PT Sumber Alfaria Trijaya Tbk's unaudited interim consolidated financial statements for the nine months ended September 30, 2020, board-authorized for issue on October 27, 2020 - just 27 days after quarter-end. Read as a nine-month blend, the headline is a mild story: net revenue grew 4.2% to Rp56,369,133 million from Rp54,110,967 million, and net income attributable to owners fell 1.8% to Rp638,403 million from Rp650,230 million - both figures the Company's own management presentation states directly.

But subtracting H1 2020's already-published figures from this filing's nine-month totals isolates Q3 standalone, and it tells a more interesting story than either the 9M blend or last quarter's uniform collapse: net revenue grew a modest 1.8% year-over-year (Rp18,283,777 million vs Rp17,953,858 million), operating income fell 32.7%, EBITDA fell 7.5%, and net income to owners fell 41.9% (Rp145,146 million vs Rp249,869 million) - profit deteriorated further even as the top line stabilized. The reason isn't the segments themselves - it's what happened above them (see Segments Compared).

Outside Java, the segment flagged as broken for the first time last quarter, recovered on both fronts: standalone Q3 revenue grew 8.8% and segment income grew 24.6%, with margin expanding to 4.69% from 4.09% a year earlier - a clean reversal of Q2's setback (see Segment Performance). Jabodetabek didn't recover: its standalone Q3 segment income fell 34.4% and its margin kept eroding, to 2.39% from 3.56%. And the reason profit still fell harder than the segments alone would suggest is that unallocated corporate overhead, which fell in step with segment income during Q2's collapse, did not fall in step in Q3 - it grew to 64.9% of combined segment income, from 51.5% a year earlier, even as combined segment income itself declined 7.0%. The overhead-discipline thread this site tracked through five straight periods broke this quarter, not last quarter.

Meanwhile the balance sheet did the opposite of the P&L: net gearing collapsed to roughly 0.03x at September 2020, from 0.57x at June 2020, as cash nearly tripled quarter-over-quarter (Rp1,112,627 million to Rp2,970,155 million) while interest-bearing debt kept falling (see Key Financial Metrics). The balance-sheet repair this site's June post flagged as a yellow watch item has, on this site's own consistent methodology, more than reversed itself in a single quarter.

The Prescription

Alfamart's task now is to figure out why unallocated overhead didn't flex down with segment income in Q3, the way it did during Q2's much sharper collapse (see Segments Compared) - management should disclose, next quarter, what specifically grew in that unallocated line (corporate headcount, IT/digital investment, one-off pandemic costs) rather than leaving readers to infer it from a residual. Outside Java's recovery (see Segment Performance) proves the underlying retail business can still grow through a pandemic; the overhead line is now the thing standing between that recovery and the bottom line actually reflecting it. What it should stop doing: continuing to grow rental-and-promotional-participation income from suppliers at a rate (+68.8% year-over-year for the nine months, Rp2,813,841 million vs Rp1,667,047 million, see Beyond the Usual) that far outpaces net revenue growth (+4.2%), without disclosing what's driving it - a supplier-facing income line growing that much faster than the business itself invites the question of whether suppliers are being squeezed harder to offset the retail margin pressure documented above, and the filing offers no explanation either way.

Key Financial Metrics

9M 2020 vs. 9M 2019 (P&L and cash flow), September 2020 vs. December 2019 (balance sheet) - consolidated, unaudited

FX: IDR 14,879.00 = USD 1 (period-end exchange rate as of September 30, 2020, the last trading day of the quarter) - a roughly 4.8% Rupiah depreciation from the Rp14,194.05 used at June 30, 2020, a partial reversal of Q2's stabilization.

Metric	9M 2020 (IDR)	9M 2020 (USD)	9M 2019 (IDR)	YoY
Net Revenue	Rp56,369,133M	~$3,788.5M	Rp54,110,967M	✅ +4.2%
Gross Profit	Rp11,482,394M	~$771.7M	Rp10,785,560M	✅ +6.5%
Income from Operations ("Operating Income")	Rp1,023,005M	~$68.8M	Rp1,123,734M	⚠️ -9.0%
Finance Cost	Rp299,558M	~$20.1M	Rp302,120M	✅ -0.8%
Income Before Final Tax and Corporate Income Tax	Rp800,710M	~$53.8M	Rp858,306M	⚠️ -6.7%
Income for the Period (total)	Rp659,186M	~$44.3M	Rp667,737M	⚠️ -1.3%
Net Income (attributable to owners)	Rp638,403M	~$42.9M	Rp650,230M	⚠️ -1.8%
EPS	Rp15.37	~$0.0010	Rp15.66	⚠️ -1.8%
EBITDA» (Operating Income + D&A)	Rp3,158,206M	~$212.3M	Rp3,174,523M	⚠️ -0.5%

Balance sheet metric	Sep 2020 (IDR)	Sep 2020 (USD)	Dec 2019 (IDR)	Change
Total Assets	Rp25,347,892M	~$1,703.3M	Rp23,992,313M	✅ +5.7%
Total Liabilities	Rp18,351,029M	~$1,233.4M	Rp17,108,006M	⚠️ +7.3%
Total Equity (attributable to owners)	Rp6,794,165M	~$456.6M	Rp6,696,944M	✅ +1.5%
Total Cash (incl. time deposits)	Rp2,970,155M	~$199.6M	Rp3,909,150M	⚠️ -24.0%

Against June 2020's balance sheet, total cash nearly tripled in a single quarter (Rp1,112,627 million to Rp2,970,155 million, +166.9%), while total assets and equity both recovered past their December 2019 starting point for the first time since Q1's short-term-borrowing spike. The Company's own presentation still frames this as Net Revenue growing 4.2% to Rp56.37 trillion and Net Profit decreasing slightly by -1.8% - the same nine-month blend this post's opening section isolates Q3 standalone from.

"Adjusted EBITDA^»" still isn't a metric Alfamart reports; the EBITDA line above is Income from Operations plus depreciation and amortization (Rp2,135,201M for 9M 2020, per the segment note, versus Rp2,050,789M in 9M 2019), matching the convention used in every prior post here. Isolating Q3 standalone (9M minus the already-published H1 2020 figures), D&A grew 7.1% year-over-year (Rp728,338M vs Rp680,264M), roughly tracking revenue growth rather than distorting the EBITDA comparison.

Free cash flow^» (operating cash flow minus capex, both from the filed cash flow statement) rose to +Rp2,760,013M for 9M 2020, from +Rp2,932,631M in 9M 2019 - a modest 5.9% decline for the nine months that masks a sharp standalone-quarter swing: isolating Q3 2020 against the already-published H1 2020 figure (-Rp649,524 million), standalone Q3 free cash flow was +Rp3,409,537 million, up 52.0% from Q3 2019's implied +Rp2,242,940 million and the strongest single quarter this site has recorded for this company. Net cash from operations for the nine months was Rp4,054,364M (from Rp3,746,604M in 9M 2019), on cash receipts from customers of Rp56,160,201M (from Rp54,604,553M) against cash payments for income taxes that grew faster still (Rp530,310M from Rp397,204M). Capex ("Perolehan aset tetap") rose 59.0% for the nine months (Rp813,973M to Rp1,294,351M), consistent with the continued-expansion posture flagged at H1 2020.

Debt-to-Equity Ratio (DER)^» (interest-bearing debt, including finance leases and consumer financing, over equity attributable to owners, excluding the right-of-use lease liability to stay on the same basis as prior posts): 0.47x at September 2020, down sharply from 0.74x at June 2020 and now below December 2019's 0.74x for the first time this site has tracked, on total interest-bearing debt of approximately Rp3,188,140 million (down from Rp4,889,766 million at June 2020), as the two 2020-maturity bonds' repayment continued to work through the balance sheet without fresh refinancing. Net gearing fell even further, to roughly 0.03x at September 2020, from 0.57x at June 2020 - a near-complete reversal of the divergence flagged as a yellow watch item last quarter (see Beyond the Usual), since this quarter cash grew and debt fell simultaneously rather than one offsetting the other. The Company's own presentation deck reports a net gearing figure of "0.17x, marginally increased from 0.03x" for the same date - a different trajectory than this site's methodology shows, reflecting a different debt scope (this site's convention excludes the right-of-use lease liability recognized under PSAK 73; including it, per the deck's likely basis, would push net debt meaningfully higher) - the same scope divergence this site has flagged in every quarter since PSAK 73 took effect.

Key Operational Metrics

• Mini-market network (Alfamart brand only): 15,102 as of September 2020 (11,254 directly owned, 3,848 under franchise agreement), up from 14,310 at December 2019 (10,614 owned, 3,696 franchised) - a net addition of 792 stores for the nine months, disclosed directly in the financial-statement notes and matching the Company's own presentation figure exactly. Isolating Q3 standalone against June 2020's already-published 14,752, Q3 alone added 350 stores - faster than both Q1's 237 and Q2's pandemic-paused 205, the fastest quarter of 2020 so far.
• Full network (Alfamart, Alfamidi, Lawson, Dan+Dan): 17,129 stores as of September 2020 (32% Greater Jakarta, 37% Java excluding Greater Jakarta, 31% Outside Java) - up from 16,720 at June 2020, with Outside Java's share of the network edging up a further point from December 2019's 30%, continuing the multi-year geographic shift the presentation itself charts back to 2016.
• Permanent employees: 66,082 as of September 2020, up 11.6% from 59,214 at December 2019 for the nine months - isolating Q3 standalone against June 2020's 63,765 shows headcount grew 3.6% in the quarter, still outpacing the quarter's 2.4% store-count growth, though by a narrower margin than earlier in the pandemic.
• Market share: Alfamart & Alfamidi's combined share of Total Indonesia grocery grew from 10.7% to 12.1% year-to-date September 2019 vs September 2020 (from 10.7% to 11.8% at the equivalent H1 mark), and their share of Modern Trade grew from 21.9% to 23.1% (from 21.5% to 22.6% at H1) - share gains that widened in Q3 even as the industry backdrop below worsened, per the presentation's Nielsen Retail Audit data.
• Industry backdrop - materially worse than H1's already-weak reading: Indonesia's total grocery trade (68 FMCG categories, Nielsen) fell 4.9% year-to-date September 2020, reversing from +4.2% a year earlier - a sharp deterioration from H1's still-positive-but-weak +0.7%, meaning Q3 alone dragged a modestly positive first-half industry reading into an outright year-to-date decline. Modern Trade specifically slowed further, to +2.0% from +7.7% a year ago; MT Minimarket slowed to +6.0% from +12.7%; Traditional Trade fell to -11.5% from +1.1%; and Super/Hyper's decline deepened to -9.5% from -4.5%. Alfamart grew revenue and gained share through an industry backdrop that got worse, not better, over the same window.
• Seasonality note: Q3 (July-September) carries no Ramadan tailwind in either year, keeping this quarter's year-over-year comparison seasonally clean, unlike H1's Ramadan-driven Q2.

Segment Performance

Alfamart reports the same three geographic segments as every prior post - Jabodetabek (greater Jakarta), Java excluding Jabodetabek, and Outside Java. Figures below use each segment's external revenue, for the nine months ended September 30.

Segment	Revenue (9M 2020)	Revenue (9M 2019)	YoY	Segment Income (9M 2020)	Segment Income (9M 2019)	YoY	Margin (2020 vs 2019)
Jabodetabek	Rp20,060,981M	Rp19,377,105M	✅ +3.5%	Rp573,958M	Rp680,112M	⚠️ -15.6%	⚠️ 2.86% vs 3.51%
Java (excl. Jabodetabek)	Rp19,566,069M	Rp19,526,384M	⚠️ +0.2%	Rp1,065,829M	Rp1,182,824M	⚠️ -9.9%	⚠️ 5.45% vs 6.06%
Outside Java	Rp16,742,083M	Rp15,207,478M	✅ +10.1%	Rp711,678M	Rp599,715M	✅ +18.7%	✅ 4.25% vs 3.94%

Read as a nine-month blend, this looks like a continuation of H1's pattern - two segments losing margin, one gaining. Isolating Q3 standalone (9M totals minus the already-published H1 2020 figures) tells a genuinely different story than either the blend above or Q2 2020's uniform collapse:

Segment	Revenue (Q3 2020, implied)	Revenue (Q3 2019, implied)	YoY	Segment Income (Q3 2020, implied)	Segment Income (Q3 2019, implied)	YoY	Margin (2020 vs 2019)
Jabodetabek	Rp6,336,837M	Rp6,466,947M	⚠️ -2.0%	Rp151,231M	Rp230,455M	⚠️ -34.4%	⚠️ 2.39% vs 3.56%
Java (excl. Jabodetabek)	Rp6,313,605M	Rp6,310,772M	⚠️ +0.04%	Rp336,292M	Rp365,302M	⚠️ -7.9%	⚠️ 5.33% vs 5.79%
Outside Java	Rp5,633,335M	Rp5,176,139M	✅ +8.8%	Rp263,955M	Rp211,937M	✅ +24.6%	✅ 4.69% vs 4.09%

Outside Java - the region FY2019, Q1 2020, and Q2 2020 all treated as this company's real growth engine, and the one segment Q2 flagged as having lost momentum for the first time - recovered cleanly this quarter: standalone revenue growth accelerated to 8.8% (from Q2's barely-positive 2.1%), segment income grew 24.6% (reversing Q2's 20.4% decline), and margin expanded 60 basis points year-over-year to a level above where it stood in Q3 2019. This is the clearest single-quarter reversal this site has recorded for any Alfamart segment.

Jabodetabek did not recover. Standalone revenue turned negative (-2.0%, worse than Q2's flat +0.1%) and segment income fell 34.4% - a deeper decline than Q2's already-sharp 24.8% drop, and now three consecutive quarters (Q1's erosion, Q2's collapse, Q3's continued fall) of margin loss for the capital region specifically.

Java excluding Jabodetabek stayed weak but stopped getting worse: standalone revenue was essentially flat (+0.04%, versus Q2's outright 11.3% decline), and while segment income still fell 7.9%, that's a materially smaller decline than Q2's 35.6% drop. This segment's trajectory reads as stabilizing, not recovering.

Segments Compared

Combined segment income fell 4.5% for the nine months (Rp2,462,651M to Rp2,351,465M), a smaller decline than H1's -3.3% would suggest given Q2's sharp standalone hit, because Q3's partial recovery (see above) offset some of Q2's damage. Isolating Q3 standalone tells the story this post's opening flags: combined segment income fell 7.0% (Rp807,694M to Rp751,478M), but unallocated corporate overhead did not fall in step - it actually rose 17.3% (Rp415,957M to Rp487,783M), pushing overhead's share of combined segment income from 51.5% in Q3 2019 standalone to 64.9% in Q3 2020 standalone. This breaks the overhead-discipline pattern FY2019 and Q2 2020 (where overhead fell 28.1% almost exactly in step with segment income's 29.2% decline) both tracked - the one quarter segment-level performance actually improved is the one quarter overhead flexed the wrong way. The filing doesn't itemize what grew inside "unallocated operating expenses," so the specific driver (see The Prescription) is a genuine open question for next quarter's filing.

Beyond the Usual

The filed financial statements still make no mention of Covid-19, three quarters into the pandemic

For a third consecutive quarter, PT Sumber Alfaria Trijaya Tbk's unaudited interim consolidated financial statements contain no mention of Covid-19, the pandemic, or any related risk anywhere in the notes - not as a risk factor, not in a subsequent-events disclosure, not anywhere. The filing was authorized for issue on October 27, 2020, 27 days after quarter-end, comparable to Q2 2020's turnaround. As with the two prior quarters, the Company's own concurrently issued management presentation discusses the pandemic's effect on the industry directly (a 4.9% year-to-date decline in total Indonesia grocery, explicitly attributed to demand conditions this filing doesn't reference at all). Three quarters running, the Company has written about the pandemic in investor communications while keeping the legally filed financial statements silent on it, even as this quarter's own segment-level swings (Outside Java's recovery, Jabodetabek's continued erosion, the overhead-discipline break documented in Segments Compared) plausibly connect to exactly the demand conditions the deck describes but the filing doesn't.

Unallocated overhead grew as a share of segment income in the one quarter segment performance actually improved

Combined segment income fell 7.0% standalone in Q3 2020, a much milder decline than Q2's 29.2% collapse - yet unallocated corporate overhead grew 17.3% in the same quarter, after having fallen almost exactly in step with segment income during Q2. Overhead's share of combined segment income jumped from 51.5% to 64.9% year-over-year. The filing discloses "unallocated operating expenses" as a single residual line item in the segment note, with no breakdown of what specifically grew - corporate headcount (up 3.6% quarter-over-quarter per Key Operational Metrics), digital/IT investment, or one-off items aren't distinguished from each other. A cost-discipline pattern this site has tracked as durable across five prior periods broke in the one quarter it mattered least to break, from a pure segment-performance standpoint - worth watching whether it's a one-quarter anomaly or a new baseline.

Supplier-funded shelf-space and promotional income grew nearly 17x faster than net revenue

Income from space rental and promotional participation fees charged to suppliers - a component of Net Revenue, not a separate line - grew 68.8% for the nine months (Rp1,667,047 million to Rp2,813,841 million), against 4.2% net revenue growth overall. This is a supplier-facing revenue stream, distinct from consumer sales, and its scale roughly quadrupled as a share of the increase in total net revenue growth for the period. The filing doesn't explain the driver - whether it reflects new promotional-space inventory, renegotiated per-supplier rates, or suppliers competing harder for shelf space as overall category growth turned negative (see the industry backdrop in Key Operational Metrics) - but the disproportion is large enough that a reader relying on the net revenue headline alone would miss where a meaningful share of this quarter's growth actually came from.

Investment in associated companies grew 83% on 4.3x higher cash injections, while losses from those associates widened

Investment in associated companies rose to Rp334,430 million at September 2020, from Rp182,677 million at December 2019 - an 83.1% increase - funded by cash payments for additional investment in associates of Rp169,378 million for the nine months, more than 4.3 times the Rp38,988 million spent in the same period last year. Over the same window, the Group's share of losses from these associates more than doubled, to Rp27,489 million from Rp11,704 million. The filing doesn't name which associate(s) received the capital or drove the wider losses, continuing a pattern this site has tracked around the insurtech-stake vehicle since FY2019 without full transparency into which associate is which.

The remaining 2018 bond series stays a current liability, unchanged from last quarter

The Company's one remaining outstanding bond - the Rp1,000,000 million "Obligasi Berkelanjutan II Sumber Alfaria Trijaya Tahap II Tahun 2018," maturing April 12, 2021 - remains classified entirely as a current liability (Rp999,597 million net of unamortized issuance cost), the same classification Q2 2020 reported. No refinancing plan has been disclosed in this filing either, five months out from maturity.

Franchise income kept falling even as the franchised store count grew further

Income from franchise agreements came to Rp240,989 million for the nine months, down 3.3% from Rp249,142 million in 9M 2019 - continuing the small decline Q2 2020 first flagged (a 1.8% H1 decline), even as the franchised portion of the Alfamart-brand network grew further, to 3,848 stores from 3,696 at December 2019 (+4.1%). A shrinking franchise-income line against a growing franchise store count, now confirmed across two consecutive filings, continues to suggest individual franchisee stores' own sales are running softer than the network's headline store-count growth would imply.

Coverage Table

Metric	9M 2020	9M 2019	YoY	Why it matters
Net Revenue (9M)	Rp56,369,133M	Rp54,110,967M	✅ +4.2%	Headline growth, but see the Q3-standalone figure below
Net Revenue (Q3 standalone, implied)	Rp18,283,777M	Rp17,953,858M	✅ +1.8%	Growth held up even as the industry backdrop worsened
Net Income to owners (Q3 standalone, implied)	Rp145,146M	Rp249,869M	⚠️ -41.9%	Profit fell faster than revenue on the overhead-discipline break
Outside Java segment income (Q3 standalone, implied)	Rp263,955M	Rp211,937M	✅ +24.6%	The growth-engine segment flagged as broken last quarter recovered cleanly
Net gearing	0.03x	0.57x (Jun 2020)	✅ down sharply	Cash tripled and debt fell in the same quarter, reversing last quarter's watch item

Target Valuation Range

Enterprise value ~Rp27.83 trillion (~$1.87B), implying 6.25x EV/EBITDA and a trailing P/E of ~25.1x - Alfamart is cheaper than it was three months ago on every multiple this site tracks. The reason is almost entirely the share price - trailing earnings barely moved, but the stock fell further than the business did. That's not obviously a buying opportunity (Jabodetabek's margin erosion and the new overhead-discipline break are real, unresolved problems), but it is a genuine de-rating, not a re-rating of risk the market had already priced in.

Alfamart's shares closed at Rp665 on September 30, 2020 (the last trading day of the quarter) - down 16.4% from Rp795 at June 2020, and the lowest quarter-end close this site has recorded for this company.

Market cap → enterprise value	Q3 2020
Share price (period-end)	Rp665
Shares outstanding	41,524,501,700
Market capitalization	Rp27.61 trillion (~$1.86B)
Plus: interest-bearing debt	Rp3.19 trillion
Less: cash and equivalents	Rp2.97 trillion
Enterprise value	Rp27.83 trillion (~$1.87B)

Peer-multiple sanity check	Q2 2020	Q3 2020	Change
Trailing P/E	~27.4x	~25.1x	✅ down
EV/EBITDA	~8.11x	~6.25x	✅ down sharply
P/B	~4.97x	~4.07x	✅ down

Trailing P/E (using trailing-twelve-month net income to owners of Rp1,100,686 million: FY2019's Rp1,112,513 million, less 9M 2019's Rp650,230 million, plus this period's Rp638,403 million) compressed as the price fell faster (16.4%) than trailing earnings did (8.7%). EV/EBITDA (against trailing-twelve-month EBITDA of Rp4,455,481 million) fell sharply, both because EV fell (a smaller net-debt add-on now that gearing has collapsed) and trailing EBITDA held roughly flat. P/B (book value of approximately Rp6,794,165 million equity attributable to owners ÷ 41.52 billion shares) fell tracking the share price decline more than any change in book value, which actually grew slightly this quarter.

No listed domestic peer still exists for a direct minimarket-format comparison - Indomaret remains privately held under the Salim Group. A full DCF still isn't included here: Q3's segment-level recovery is a genuinely different signal than Q2's uniform collapse, but the overhead-discipline break documented in Segments Compared is new and unexplained, and one more quarter is needed to see whether it reverses or becomes the new baseline before a forward cash-flow model would mean anything more than false precision. The share price moved within a Rp665-Rp1,025 range across the trailing two years to this quarter-end - a 35.1% peak-to-trough swing, above the threshold this site treats as warranting explicit attention - with September's close sitting at the low end of that range and the lowest point of the entire window, a continuation of the steady monthly decline that began after April 2020's Rp870 local high.

---