AMRT Did the Margin Recovery Survive Without Ramadan?

The Real Test Comes Back Clean

The ninth post on this company found Jabodetabek and Java excluding Jabodetabek both losing segment margin in Q1 2019. The tenth, covering H1 2019, found that erosion reversed almost entirely - but flagged the reversal as suspect, since it landed in the exact quarter Ramadan's fasting-month spending fell (May-June 2019), and closed by naming Q3 2019 as "the real test" of whether the underlying two-region weakness had actually been fixed or simply gone into hiding for two months. This is PT Sumber Alfaria Trijaya Tbk's unaudited interim consolidated financial statements for the nine months ended September 30, 2019 - a quarter with no Ramadan tailwind at all - and the test comes back clean: net revenue grew 9.1% to Rp54,110,967 million from Rp49,605,115 million, income from operations rose 32.1% to Rp1,123,734 million from Rp850,378 million, and net income attributable to owners jumped 94.0% to Rp650,230 million from Rp335,167 million. EPS rose from Rp8.07 to Rp15.66.

Isolating the implied Q3-only figures (this filing's nine-month total less the already-reported H1 2019 numbers) answers the H1 post's question directly: every one of the three segments still gained margin year-over-year in the quarter that carried none of Ramadan's lift. Jabodetabek's implied Q3-only margin rose to roughly 3.56% from about 3.34% a year earlier, Java excluding Jabodetabek's to roughly 5.79% from about 4.83%, and Outside Java's to roughly 4.09% from about 3.20% - smaller gains than the Ramadan-boosted Q2 swing H1 2019 reported, but genuine gains all the same, in the one quarter this year with nothing seasonal doing the work. Q1's margin erosion wasn't a structural problem masked by a calendar effect - it was a real, if temporary, dip that the business has since worked through.

The Prescription

Alfamart's real opportunity now is to lean harder into Outside Java, which has posted the largest nine-month cumulative margin gain of the three segments for a third straight post (135 basis points this time, on segment income growth of 74.6%) while still growing revenue fastest of the three regions (+14.9%) - it's the one segment where growth and margin expansion are compounding together rather than trading off against each other, and it's still catching up to the other two regions' absolute margin level, meaning there's more room to run before it hits the same maturity ceiling Jabodetabek and Java excluding Jabodetabek appear to be approaching. What it should stop doing: letting permanent headcount keep outgrowing the store network without explaining why in its own disclosures - permanent employees grew 11.4% from December 2018 (51,656 to 57,559) against a store count that grew only about 2.9% over the same nine months, continuing a mismatch this site has now flagged across multiple periods since H1 2018, and it shows up directly in selling and distribution expenses growing 10.4% - faster than revenue's 9.1% growth. A management team that just proved its segment margins can hold up without a seasonal crutch shouldn't let unexplained overhead growth quietly erode the same gains from the cost side.

Key Financial Metrics

9M 2019 vs. 9M 2018 (P&L and cash flow), and September 2019 vs. Dec 2018 (balance sheet) - consolidated, unaudited

FX: IDR 14,219.4 = USD 1 (a validated period-end exchange rate as of September 30, 2019, the last day of the quarter - used for every USD conversion below).

Metric	9M 2019 (IDR)	9M 2019 (USD)	9M 2018 (IDR)	YoY
Net Revenue	Rp54,110,967M	~$3,805.4M	Rp49,605,115M	✅ +9.1%
Gross Profit	Rp10,785,560M	~$758.5M	Rp9,698,486M	✅ +11.2%
Income from Operations ("Operating Income")	Rp1,123,734M	~$79.0M	Rp850,378M	✅ +32.1%
Finance Cost	Rp302,120M	~$21.2M	Rp411,363M	✅ -26.6%
Income Before Tax	Rp858,306M	~$60.4M	Rp436,222M	✅ +96.8%
Income for the Period (total)	Rp667,737M	~$47.0M	Rp344,527M	✅ +93.8%
Net Income (attributable to owners)	Rp650,230M	~$45.7M	Rp335,167M	✅ +94.0%
EPS	Rp15.66	~$0.0011	Rp8.07	✅ +94.0%
EBITDA» (Operating Income + D&A)	Rp3,174,523M	~$223.3M	Rp2,787,684M	✅ +13.9%

Balance sheet metric	Sept 2019 (IDR)	Sept 2019 (USD)	Dec 2018 (IDR)	Change
Total Assets	Rp23,255,203M	~$1,635.4M	Rp22,165,968M	✅ +4.9%
Total Liabilities	Rp16,682,297M	~$1,173.2M	Rp16,148,410M	⚠️ +3.3%
Total Equity (attributable to owners)	Rp6,393,347M	~$449.6M	Rp5,856,468M	✅ +9.2%
Total Cash	Rp3,411,280M	~$239.9M	Rp2,070,429M	✅ +64.8%

"Adjusted EBITDA^»" still isn't a metric Alfamart reports; the EBITDA line above is Income from Operations plus depreciation and amortization (Rp2,050,789M for 9M 2019, per the segment note, versus Rp1,937,306M in 9M 2018), matching both the company's own presentation-deck figure and the convention used in every prior post here.

Free cash flow^» (operating cash flow minus capex, both from the filed cash flow statement) fell to +Rp2,932,631M for 9M 2019, from +Rp3,432,510M in 9M 2018 (-14.6%) - net cash from operations fell 5.0% (Rp3,942,841M to Rp3,746,604M) while capex ("Perolehan aset tetap") rose 59.5% (Rp510,331M to Rp813,973M), extending the network-investment resumption H1 2019 already reported as a multi-quarter pattern rather than a one-off. Total cash still rose a sharp 64.8% from December 2018 to Rp3,411,280M, a reversal of H1's 18.0% decline - operating cash generation plus a much lighter financing outflow this quarter (Rp568,112M net used, against Rp2,389,612M in the same nine months of 2018, when the Company was still repaying a maturing bond series) rebuilt the cash balance despite the heavier capex pace.

Debt-to-Equity Ratio (DER)^» (interest-bearing debt, including finance leases, consumer financing, and the JD.com borrowing discussed at FY2018, over equity attributable to owners, computed directly from this filing's own balance-sheet columns): 0.81x at September 2019, down from 0.85x at June 2019 and 0.90x at December 2018 - a fourth consecutive period of deleveraging, on total interest-bearing debt of approximately Rp5,150,381 million (from Rp5,265,620 million at December 2018). Net gearing (net debt over the same equity base) fell sharply to 0.27x, from 0.57x at June 2019, almost entirely on this quarter's cash rebuild rather than debt repayment - gross interest-bearing debt barely moved. The Company's own narrower covenant metric (interest-bearing debt excluding the JD.com borrowing, over equity) came in at 0.71x, down from 0.75x - both this and this site's own broader figure stay comfortably inside the Company's 2.5x bond covenant, which the filing confirms was met. The Company's own presentation reports a materially better net gearing ratio of 0.17x, down from 0.87x a year earlier (YTD Sep 2018) - a different net-debt definition and comparison base than this site's own DER calculation above.

ROAA^» / ROAE^» (nine-month, not annualized, using net income attributable to owners over average total assets and average equity attributable to owners, the same basis as every prior post here): approximately 2.86% / 10.62% for 9M 2019 - the Company's own presentation shows a closely comparable 2.86% / 10.33% for the same period.

Key Operational Metrics

• Total store network: 15,742 stores as of September 2019 (13,996 Alfamart, including 329 Alfaexpress; 1,520 Alfamidi; 52 Lawson; 174 Dan+Dan), up from 15,539 at June 2019 - a net addition of 203 stores in the quarter, and up from 15,294 at December 2018 (448 net additions over the nine months).
• Permanent employees: 57,559 as of September 2019, up 11.4% from 51,656 at December 2018 and up 5.6% from 54,481 at June 2019 - continuing to outgrow the roughly 2.9% store-count growth over the same nine months by an even wider margin than H1 2019 already reported (see The Prescription above).
• Warehouses: 43 nationwide as of September 2019 (32 for Alfamart, 10 for Alfamidi, 1 for Dan+Dan), the first change since September 2017, when the count had been unchanged at 42 across every period in between.
• Market share: the combined Alfamart/Alfamidi share of Indonesia's minimarket-format modern-trade channel fell again to 34.7% for the nine months ended September 2019, from 36.4% in the same nine months of 2018 (the Company's own presentation shows Alfamart's share of the channel alone at 29.7%, down from 31.1%, and Alfamidi's at 5.0%, down from 5.3%) - a third consecutive period of decline in this metric, continuing the reversal H1 2019 first flagged.
• Seasonality note: this quarter carries no Ramadan effect - the fasting month fell entirely within Q2 2019 (see H1 2019) - making the segment margin gains reported above the cleaner read on the underlying business (see The Real Test Comes Back Clean above).
• Industry backdrop: modern trade grew 7.6% for the nine months (from 3.7% a year earlier, per the Company's own Nielsen-sourced data), again driven by the minimarket format accelerating to 12.9% growth (from 7.0%), while traditional trade slipped to a small decline (-1.1%, from +1.0%) and super/hyper's decline deepened further (-5.8%, from -3.9%). Alfamart's own market-share decline within that still-growing minimarket channel (above) means the gap between industry growth and the Company's own share continues to widen.

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 (excluding the minor inter-segment elimination Jabodetabek reports).

Segment	Revenue (9M 2019)	Revenue (9M 2018)	YoY	Segment Income (9M 2019)	Segment Income (9M 2018)	YoY	Margin (2019 vs 2018)
Jabodetabek	Rp19,377,105M	Rp18,331,578M	✅ +5.7%	Rp680,112M	Rp570,340M	✅ +19.2%	✅ 3.51% vs 3.11%
Java (excl. Jabodetabek)	Rp19,526,384M	Rp18,041,354M	✅ +8.2%	Rp1,182,824M	Rp934,596M	✅ +26.6%	✅ 6.06% vs 5.18%
Outside Java	Rp15,207,478M	Rp13,232,183M	✅ +14.9%	Rp599,715M	Rp343,446M	✅ +74.6%	✅ 3.94% vs 2.60%

Jabodetabek posted the smallest margin gain of the three on a nine-month cumulative basis (40 basis points to 3.51%), consistent with it being the most mature of the three regions. But the number that actually answers this post's question sits in the implied Q3-only slice: subtracting H1 2019's already-reported figures shows Jabodetabek's Q3-only margin at roughly 3.56%, up from an implied ~3.34% in Q3 2018 - a genuine, if modest, improvement in the one quarter with no seasonal help.

Java excluding Jabodetabek again posted the highest nine-month margin of the three (6.06%) and the widest absolute margin gain (88 basis points). Its implied Q3-only margin of roughly 5.79% compares to an implied ~4.83% a year earlier - the largest Q3-only improvement of the three segments, and notably larger than Jabodetabek's, suggesting this region's underlying recovery is running ahead of Jabodetabek's.

Outside Java again grew revenue fastest (+14.9%) and posted by far the largest nine-month margin improvement (135 basis points, on segment income up 74.6%) - extending the turnaround every prior post this year has reported. Its implied Q3-only margin of roughly 4.09% compares to an implied ~3.20% a year earlier, the second-largest Q3-only gain of the three segments (behind Java excluding Jabodetabek's) but consistent with a region still closing the gap on Jabodetabek and Java excluding Jabodetabek's margin levels rather than having caught up.

Isolating the implied Q3-only figures this way (this filing's nine-month total less the already-reported H1 2019 figures) is what actually settles the question H1 2019 raised: all three regions still gained margin year-over-year in the one quarter that carried no Ramadan tailwind, confirming the H1 swing reflected a real recovery from Q1's erosion rather than a seasonal illusion layered on top of unresolved weakness.

Segments Compared

Combined segment income rose 33.2% (Rp1,848,382M to Rp2,462,651M) for the nine months - faster than both consolidated operating income growth (32.1%) and net revenue growth (9.1%), continuing the pattern of all three segments gaining margin together. Unallocated corporate overhead grew 34.2% (Rp998,004M to Rp1,338,917M), essentially keeping pace with combined segment income growth rather than outrunning it - overhead's share of combined segment income ticked up only slightly, from 54.0% to 54.4%, a far smaller shift than the roughly 260-basis-point jump H1 2019 reported for the half. Overhead discipline isn't actively improving profit growth this quarter the way it did at Q1 2019, but it also isn't actively working against it the way H1's read suggested - the mix has roughly stabilized.

Beyond the Usual

Alfamidi's original Lawson license ended in January 2019, replaced by a dedicated subsidiary's own 18-year agreement

Subsidiary MIDI's Master License Agreement with Lawson, Inc. of Japan - signed June 20, 2011, granting the exclusive right to sub-franchise the Lawson brand in Indonesia - formally ended on January 7, 2019. It had already been superseded on June 6, 2018 by a separate 18-year Master License Agreement between Lawson Japan and a different subsidiary, LWS, created specifically to hold the Lawson brand relationship going forward; LWS pays Lawson a royalty based on store-level net revenue. This restructuring coincides with a genuine acceleration in the brand's own store count - Lawson stores grew from 44 at H1 2019 to 52 this quarter, the fastest single-quarter growth of any brand in the portfolio - suggesting the license was moved into a dedicated entity ahead of a deliberate push to grow the format, not as a routine renewal.

A further capital injection into the insurtech stake's holding vehicle, right after quarter-end

On October 1, 2019 - a subsequent event disclosed in this filing - PT Sumber Trijaya Lestari (STL), the subsidiary that paid Rp14,000 million for a 14.28% stake in online insurance platform PT Kita Indonesia Plus as a subsequent event at H1 2019, had its own issued capital increased by a further Rp30,000 million (Rp29,985 million from the Company, Rp15 million from subsidiary MIDI, keeping the 99.95%/0.05% split unchanged). The filing doesn't disclose what the fresh capital is earmarked for, but the timing - one day after this quarter's period end, into the same vehicle that made the insurtech investment three months earlier - reads as continued funding appetite for that line of business rather than a one-off stake purchase, worth tracking at the next quarter for what STL actually does with it.

Franchise income growth normalized sharply after H1's Ramadan-driven spike

Income from franchise agreements, embedded within net revenue, rose to Rp249,142 million for 9M 2019 from Rp196,740 million in 9M 2018 - a 26.6% increase, still ahead of the 9.1% growth in consolidated net revenue but well below the 35.6% growth H1 2019 reported for the half alone. Isolating the implied Q3-only franchise income (Rp81,136 million, against an implied ~Rp72,876 million a year earlier) shows growth slowing to roughly 11.3% for the quarter - much closer to the pace of store-network growth than H1's outsized gain, consistent with the H1 post's read that an existing franchise base paying more per store, rather than a wave of new signings, is what's actually driving the line. Unearned franchise revenue grew only 1.5% over the same nine months (Rp65,964 million to Rp66,932 million), reinforcing that reading.

The two 2020-maturity bond series still sit in current liabilities, alongside a third series not previously detailed here

This filing's bonds-payable note shows three outstanding series, not two: the Rp400,000 million 2015 Series B and Rp1,000,000 million 2017 Bonds Phase I - both maturing in May 2020 and both still classified as current liabilities this quarter, unchanged in amount from H1 2019 - plus a Rp1,000,000 million 2018 Bonds Phase II series that isn't due until April 12, 2021 and correctly stays classified as long-term. Combined nominal value across all three is Rp2,400,000 million, essentially flat from December 2018 (Rp2,393,957 million net), and nothing in this filing addresses refinancing plans for the two 2020 maturities - the same gap flagged last quarter.

Target Valuation Range

Enterprise value ~Rp43.06 trillion (~$3.03B), implying 9.75x EV/EBITDA and a trailing P/E of ~42.8x - a sixth straight period of P/E compression, but for the first time against an earnings base this post can call clean rather than seasonally inflated. The H1 2019 post deferred a fuller valuation view until this quarter clarified the underlying run rate; it has, and the read is constructive.

Alfamart's shares closed at Rp995 on September 30, 2019 (the last trading day of the quarter) - up 8.7% from Rp915 at June 2019, the first quarter-over-quarter gain since the Rp1,025 intra-window high reached in April 2019. The Company's only stock split (10-for-1, July 2013) predates this period and remains reflected in both the filed share count and today's historical price data - no further split has occurred since (confirmed against public split-history records, used only to validate that no adjustment is required here).

Market cap → enterprise value	Q3 2019
Share price (period-end)	Rp995
Shares outstanding	41,524,501,700
Market capitalization	Rp41.32 trillion (~$2.91B)
Plus: interest-bearing debt	Rp5.15 trillion
Less: cash and equivalents	Rp3.41 trillion
Enterprise value	Rp43.06 trillion (~$3.03B)

Peer-multiple sanity check	H1 2019	Q3 2019	Change
Trailing P/E	~45.6x	~42.8x	✅ down
EV/EBITDA	~9.70x	~9.75x	➖ essentially flat
P/B	~6.19x	~6.46x	⚠️ up

Trailing P/E (~42.8x, using TTM net income to owners of Rp965,201 million) is down from H1 2019 - now on a rising price against an even larger jump in trailing earnings. EV/EBITDA (against TTM EBITDA of Rp4,416,819 million) is essentially flat, as the larger EBITDA base offset the higher enterprise value. P/B (book value of ~Rp6,393,347 million equity attributable to owners ÷ 41.52 billion shares) rose, as the 8.7% price gain this quarter outpaced book value's 3.9% growth.

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, but for a different reason than H1 2019 gave: this quarter's earnings are no longer entangled with a seasonal effect, so the deferral going forward is simply that building a reliable multi-year cash-flow forecast needs more quarters of this now-cleaner run rate than this site has recorded yet, not a data-quality concern about the base year itself. The share price moved within a Rp570-Rp1,025 range across the trailing two years (a roughly 79.8% peak-to-trough swing, above the threshold this site treats as warranting its own section), with the September 2019 close sitting inside that range rather than testing either edge - the mildest two-year price picture this site has reported for AMRT since before the 2018 volatility began.

---