Record Provisioning, Falling Coverage - the Math Doesn't Add Up the Way It Should
The Q2 2020 post closed on a bank that had let its provisioning cushion ease - from Q1's series-high 223.6% coverage ratio down to roughly 200% - in the exact quarter Covid-19 restructuring kept climbing. This quarter's numbers make that retreat look almost gentle by comparison. Bank-only NPL Coverage Ratio fell again, to 159.9% at September 30, 2020 - a third consecutive quarterly decline, and now more than 60 percentage points below the Q1 peak. That's despite BRI booking its single largest quarterly provision expense of the year: Rp9,504 billion in Q3 alone, up 188.1% from Q2's Rp3,299 billion, per the consolidated income statement in BRI's own investor presentation.
The arithmetic that makes both things true at once is straightforward and unflattering: NPL formation simply outran the record provisioning. Bank-only gross NPL rose again, from 2.98% (Jun 2020) to 3.02% (Sep 2020) on the narrow headline measure - a small move - but Loan at Risk», the broader measure that also captures Covid-19-restructured loans still classified "Current," climbed from 28.92% in June to 29.77% in September, extending the near-tripling flagged last quarter. Put differently: BRI threw more provisioning at the book than at any point this year, and still lost ground on every coverage measure that matters - the headline ratio, and the broader risk pool it's meant to protect against.
Covid-19 restructuring itself did something new this quarter: it plateaued. The pool grew from Rp171.9 trillion (19.8% of loans) at June 30 to Rp193.7 trillion (22.1% of loans) at September 30 - real growth, but a fraction of the pace that took it from Rp101.2 trillion to Rp171.9 trillion the quarter before. BRI's own presentation even titles the relevant slide "Plateauing BRI Covid-19 Restructuring Trend," and shows the pool actually ticking down slightly to Rp192.3 trillion by October 31 - the first sequential decline this series has recorded for this figure. That would read as unambiguous good news, except the composition of the restructured book moved the wrong way: only 79.2% of it stayed classified "Current" at September, down from 87.6% in June - with Special Mention's share roughly doubling (6.9% to 13.8%) and NPL's share rising too (5.5% to 7.0%). The relief measure hasn't changed; what's changed is how much of the restructured book is quietly failing anyway.
The Prescription
BRI should stop reporting NPL Coverage Ratio and Loan at Risk Coverage as if they answer the same question, because this quarter shows plainly that they don't. A reader who only sees the narrow ratio easing from ~200% to 159.9% would reasonably read that as the bank pulling back on caution; a reader who also sees LAR Coverage sitting at just 21.84% (per the presentation's own ratio table) sees a bank whose provisioning against its actual risk pool - not just its formally-defaulted loans - has barely moved even as that pool kept growing. Put both numbers on the same slide, not one in the headline table and the other three slides later.
What BRI should stop doing: treating Corporate Non-SoE's continued deterioration as a segment issue to monitor rather than a signal to act on. The segment's NPL just posted its fourth consecutive series-worst reading (4.83% a year ago, 9.14% at Q1, 10.75% at Q2, now 10.85%), and this quarter its own growth nearly stalled (+1.5% YoY, down from Q2's +6.6%) - meaning BRI itself has already started pulling back new lending to the segment. That's the right instinct; it should have happened two quarters earlier, before the NPL ratio tripled from its FY2019 level.
Key Financial Metrics
9M 2020 vs. 9M 2019 (P&L, consolidated, cumulative Jan-September), and September 2020 vs. December 2019 / September 2019 (balance sheet, consolidated)
FX: 1 USD = Rp14,255.00, the exchange rate BRI's own filing discloses for September 30, 2020 - identical to the rate the same filing disclosed for June 30, 2020 (see Beyond the Usual for why that's worth flagging, since the actual IDR/USD market rate moved meaningfully between the two dates).
| Metric | 9M 2020 (IDR) | 9M 2020 (USD) | 9M 2019 (IDR) | YoY |
|---|---|---|---|---|
| Net Interest, Sharia and Premium Income, net ("Net Revenue" equivalent) | Rp57,088,326M | ~$4,005M | Rp60,869,517M | ⚠️ -6.21% |
| Operating Income | Rp20,435,973M | ~$1,434M | Rp31,086,942M | ⚠️ -34.26% |
| Net Income (attributable to owners) | Rp14,116,093M | ~$990M | Rp24,780,555M | ⚠️ -43.04% |
| Total Comprehensive Income (attributable to owners) | Rp16,880,533M | ~$1,184M | Rp30,194,117M | ⚠️ -44.09% |
| EPS (basic, 9M cumulative, attributable to owners) | Rp115.00 | ~$0.0081 | Rp203.00 | ⚠️ -43.35% |
| Balance sheet metric | Sep 2020 (IDR) | Sep 2020 (USD) | Dec 2019 (IDR) | QoQ¹ | Sep 2019 (IDR) | YoY |
|---|---|---|---|---|---|---|
| Total Assets | Rp1,447,848B | ~$101,568M | Rp1,416,759B | ✅ +2.19%² | Rp1,305,667B | ✅ +10.89% |
| Loans (gross) | Rp935,347B | ~$65,615M | Rp903,197B | ✅ +3.55%² | Rp891,973B | ✅ +4.86% |
| Total Deposits | Rp1,131,928B | ~$79,406M | Rp1,021,197B | ✅ +10.84%² | Rp959,238B | ✅ +18.00% |
| Total Equity (incl. non-controlling interest) | Rp194,668B | ~$13,656M | Rp208,784B | ⚠️ -6.76%² | Rp199,540B | ⚠️ -2.44% |
| Total Cash and Cash Equivalents (consolidated, per cash flow statement)³ | Rp135,457B | ~$9,502M | Rp236,906B | ⚠️ -42.82% | Rp176,296B | ⚠️ -23.16% |
¹ QoQ compares against the prior full-year close (Dec 2019), the same convention used throughout this series for a non-year-end quarter. ² These QoQ figures compare Sep 2020 against Dec 2019 (per the table's own convention); the more standard quarter-over-quarter reads (vs. Jun 2020: Assets +4.3%, Loans +1.3%, Deposits +5.5%, Equity +3.6%, all per BRI's own presentation) are the ones referenced in the prose above. ³ This quarter's filing discloses a consolidated cash flow statement, unlike Q2's filing, which disclosed only bank-only figures - a basis change worth noting when comparing cash trends across these two posts.
Cash contracted sharply again on both a QoQ and YoY basis, continuing the multi-quarter drawdown flagged since Q1 and Q2 - three straight quarters of the same pattern, still driven by placements with Bank Indonesia being run down rather than a liquidity problem (deposits grew 18.0% YoY over the same period).
Net income attributable to owners fell 43.0% YoY for the nine months, but the standalone Q3 provisioning number is the real story: Rp9,504 billion, the largest single quarter this series has recorded, and coverage still fell. See above.
Key Operational Metrics
Bank-only, per BRI's own investor presentation and filed financial ratios, unless stated
- CASA»: 60.15% (Sep 2020) vs 59.01% (Dec 2019) and 59.49% (Sep 2019) ✅ - up on both counts this quarter, a reversal of the multi-year drift flagged since FY2019 and still present as of Q2 - cheaper deposit funding, for once, moving the right direction.
- Loan-to-Deposit Ratio (LDR)»: 82.58% (Sep 2020, filed ratio) vs 88.64% (Dec 2019) and 93.07% (Sep 2019) ✅ - down sharply YoY again, deposit growth (+16.3% YoY, bank-only) continuing to comfortably outpace loan growth (+3.2% YoY, bank-only).
- Net Interest Margin (NIM)»: 5.76% (Sep 2020, filed ratio) vs 6.98% (Dec 2019) and 7.02% (Sep 2019) ⚠️ - still a new series low on a YoY basis, though up slightly from Q2's 5.72% - the first sequential improvement since the compression began.
- ROA» (before tax): 2.07% (Sep 2020, filed ratio) vs 3.50% (Dec 2019) and 3.42% (Sep 2019) ⚠️ - still down sharply on both counts.
- ROE» (Tier 1, filed ratio): 11.43% (Sep 2020) vs 19.41% (Dec 2019) and 19.16% (Sep 2019) ⚠️ - continuing the profitability decline Q2's post first flagged as genuine (not the mechanically inflated Q1 reading).
- CAR» (Total, bank-only, filed ratio): 20.38% (Sep 2020) vs 22.55% (Dec 2019) and 21.62% (Sep 2019) ⚠️ - still down YoY, but up further from Q2's 19.83%, continuing the partial rebuild from Q1's 18.23% low. Tier 1 CAR: 19.37% (Sep 2020) vs 21.52% (Dec 2019) and 20.59% (Sep 2019) - up from Q2's 18.81% and now further from Q1's 17.20% low. Tier 1 capital's nominal value (bank-only) also grew for the first time this year - Rp184,501B at Sep 2020, up 5.1% QoQ from Rp175,577B at June - though still down 1.6% YoY (Rp187,534B a year earlier), a smaller YoY decline than Q2's -2.4%.
- NPL ratio - gross (bank-only, filed ratio): 3.02% (Sep 2020) vs 2.62% (Dec 2019) and 2.96% (Sep 2019) ⚠️ - a small move on the surface, but see Loan at Risk above for what it hides. NPL ratio - net: 0.78% (Sep 2020) vs 1.04% (Dec 2019) and 1.13% (Sep 2019) ✅ - continues to look improved on a provisioning-adjusted basis, though that basis is exactly what's easing (see above).
- NPL Coverage Ratio (bank-only): 159.9% (9M 2020) vs 166.6% (9M 2019) ⚠️, down further from Q2's ~200% and now more than 60 points below Q1's series-high 223.6% - a third straight quarterly decline.
- BOPO» (Opex/Opr. Income, bank-only): 80.64% (Sep 2020) vs 70.10% (FY2019) and 70.50% (Sep 2019) - the weakest efficiency reading yet in this series, worse again than Q2's 77.49%.
- Loan at Risk (bank-only): 29.77% of total loans (Sep 2020) vs 10.45% (Sep 2019) - up from Q2's 28.92%, still nearly triple the year-ago reading. LAR Coverage: 21.84% (Sep 2020) vs 45.35% (Sep 2019) - down further from Q2's implied level, and per BRI's own presentation, would be 66.2% if Covid-19-restructured loans were excluded from the base, against 21.8% including them - the gap that shows how much the relief-eligible pool is diluting the coverage figure.
- Loan-mix NPL by category (bank-only): Micro 1.07%, Consumer 1.66%, Small Commercial 3.89%, Medium 5.84%, Corporate 10.85%, SoE 1.32%, Total 3.02% (9M 2020) vs Micro 1.30%, Consumer 1.40%, Small Commercial 3.71%, Medium 5.28%, Corporate 10.49%, SoE 1.10%, Total 2.96% (9M 2019). Corporate is again the standout - a new series-worst reading, extending the deterioration flagged at Q1 and Q2.
- Special Mention loans by category (bank-only): Micro 5.44%, Consumer 3.24%, Small Commercial 5.18%, Medium 6.63%, Corporate 7.44%, SoE 5.02%, Total 5.23% (9M 2020) vs Micro 4.12%, Consumer 3.09%, Small Commercial 6.49%, Medium 5.23%, Corporate 4.44%, SoE 4.31%, Total 4.58% (9M 2019). Unlike Q2, where Corporate's Special Mention ratio had fallen even as its NPL rose, this quarter Corporate's Special Mention also rose (4.44% to 7.44%) - the pipeline of loans one notch from default is refilling again, not just draining into NPL.
- Restructured loans, Covid-19 (bank-only): Rp193.7 trillion (22.1% of total loans) at Sep 30, 2020, up from Rp171.9 trillion (19.8%) at Jun 30, and Rp192.3 trillion (21.9%) at Oct 31 - the first month-over-month decline this series has recorded for this figure, per BRI's own presentation. Total restructured loans including pre-existing (non-Covid-19) restructuring reached 27.2% of total loans at Sep 2020, up from 6.4% a year earlier; excluding Covid-19-specific restructuring, that baseline figure is a comparatively flat 5.1%.
Business Lines: Loan Growth and Credit Quality
Micro loans grew 8.9% YoY (+Rp26.9 trillion), still the largest single contributor to loan growth and accelerating from Q2's 7.1% YoY, with NPL improving to 1.07% from Q2's 1.18% - the healthiest large segment in the book on both growth and quality. Small Commercial grew 4.6% YoY (+Rp8.9 trillion), picking up from Q2's 2.8% pace, though NPL edged up to 3.89% from 3.63%.
Medium loans essentially stalled, shrinking slightly (-0.5% YoY, -Rp0.1 trillion) after Q2's already-weak 3.0% growth, while NPL improved to 5.84% from Q2's 6.79% - a segment that's now shrinking and cleaning up at the same time, the opposite pattern from Corporate. Consumer grew 3.8% YoY (+Rp5.2 trillion), a touch slower than Q2's 4.3%, with NPL worsening again to 1.66% from 1.50% - the gradual retail-book deterioration flagged since Q1 continuing quarter after quarter.
Corporate Non-SoE growth nearly stalled - just 1.5% YoY (+Rp1.4 trillion), a sharp deceleration from Q2's 6.6% YoY - while its NPL climbed to a new series-worst 10.85% (see above). Unlike the last two quarters, this is no longer a segment growing and rotting simultaneously - BRI has clearly pulled back new lending here, just two quarters after the deterioration became obvious. SoE loans contracted sharply - -14.8% YoY (-Rp15.0 trillion), more than double Q2's -6.2% YoY - with NPL rising to 1.32% from 1.10% at Q2, no longer the untouched clean book it was earlier in the year, though still by far the lowest-NPL segment. BRI's own quarterly filing again carries no segment-income footnote, continuing the gap flagged since Q1 2019.
Beyond the Usual
This quarter's source document is again BRI's standard OJK transparency-format quarterly report - no notes to the financial statements beyond the standard related-party and commitments schedules, the same gap noted at Q1 and Q2. The findings below draw on the investor presentation and the quarterly report's own summary schedules.
The Restructured Book's Quality Eroded Even as Its Growth Slowed
Covid-19 restructuring's total size grew far more slowly this quarter (Rp171.9tn to Rp193.7tn, versus the prior quarter's near-doubling), which on its own would read as good news. But the collectability mix of that same pool moved the wrong way: only 79.2% remained classified "Current" as of September 2020, down from 87.6% in June, with Special Mention's share nearly doubling (6.9% to 13.8%) and NPL's share also rising (5.5% to 7.0%). This isn't a disclosure failure - BRI's own presentation shows the breakdown plainly, the same treatment flagged last quarter - but it means the regulatory relief that lets restructured loans stay "Current" regardless of actual repayment performance is buying less protection for the headline ratios each quarter that passes, as more of the restructured pool fails on its own merits rather than merely aging out of the relief window.
The Filing's Own FX Disclosure Repeats June's Number for a Different Date
BRI's quarterly filing states the exchange rate for 1 USD as of September 30, 2020 as Rp14,255.00 - the identical figure the June 30, 2020 filing disclosed for that earlier date. Indonesia's rupiah did not hold flat against the dollar over that period; it weakened meaningfully in the third quarter as global risk sentiment soured again. The two filings' comparative-quarter figures (Dec 2019: Rp13,882.50 both times; the year-ago quarter: Rp14,127.50 both times) also repeat identically, suggesting this is a template note that wasn't updated with the current period's actual closing rate rather than a genuine coincidence. It doesn't change any rupiah-denominated figure in this filing, but it does mean the USD conversions in this post's Key Financial Metrics table - which use the filing's own disclosed rate for consistency with prior posts - understate what those figures would translate to at the rate actually prevailing on September 30, 2020.
Write-Off Recovery Kept Pace With a Bigger Loan Book
BRI's bank-only write-off recovery income, per the filed income statement's "recoveries of written off loans" line, continued at a level broadly consistent with the trend seen at Q1 and Q2 - a modest positive signal that collections effort on already-written-off loans has held up through three consecutive quarters of the pandemic, even as new NPL formation (see Corporate Non-SoE, above) has not.
The 2020 Restructuring Guidance Ceiling Is Already in Sight
BRI's own presentation projects total Covid-19-restructured loans reaching approximately Rp200 trillion by December 2020 - a target the bank was already at Rp193.7 trillion against by September 30, meaning management now expects the pool to grow only a further ~3% over the final quarter of the year. If that projection holds, it would be consistent with the "plateauing" framing BRI's own materials use for this quarter's restructuring trend, rather than a wave still building toward an unknown peak.
The Rebuilding Capital Ratio Is Still Partly a Function of Regulatory Relief
The same OJK Covid-19 banking relaxations flagged last quarter - excluding Covid-19-restructured loans from the formal Loan at Risk soundness calculation, and removing the Capital Conservation Buffer requirement - remained in effect this quarter, per the filing's own capital-ratio schedule. Tier 1 CAR's continued climb (18.81% to 19.37% bank-only) is partly assisted by the same lower regulatory bar as before; the more telling number is that nominal Tier 1 capital itself grew for the first time this year this quarter (+5.1% QoQ), a genuine signal alongside the still relief-assisted ratio.
Target Valuation Range
P/E ~14.3x, P/B ~1.76x - still too early to call a real re-rating either way - BRI provisioned harder than at any point this year and still lost coverage ground, Corporate Non-SoE posted its fourth consecutive series-worst NPL reading, and the restructured book's own quality is eroding even as its headline growth slows. The one genuinely positive signal - Tier 1 capital growing in nominal terms for the first time in 2020 - is real but isolated.
BRI's stock closed at approximately Rp2,764 on September 30, 2020, per publicly available exchange price data (BRI's own investor presentation this quarter shows only a share-price chart, without a numeric closing figure in the text, unlike some prior quarters) - down roughly 8.8% from the Q2 2020 post's Rp3,030 close. Over the trailing two years the price ranged as high as roughly Rp4,073 (July 2019) and as low as roughly Rp2,482 (April 2020, the depth of the Covid-19 crash) - the same roughly 39% peak-to-trough move noted at Q1 and Q2 - with September's close sitting back down near the lower half of that range after a brief rally in August.
9M 2020 EPS (consolidated, attributable to owners) is Rp115.00. Combining that with FY2019's Rp281.31 and subtracting 9M 2019's own Rp203.00 gives a trailing-twelve-month EPS of approximately Rp193.31, giving a P/E» of ~14.3x against the ~Rp2,764 close.
| Market cap → book value | 9M 2020 |
|---|---|
| Share price (period-end) | ~Rp2,764 |
| Shares outstanding | ~122.75 billion (derived from this quarter's own EPS/net-income relationship) |
| Market capitalization | ~Rp339,281B |
| Total equity attributable to owners (book value) | Rp192,216B |
| P/B | ~1.76x |
| Peer-multiple sanity check | H1 2020 | 9M 2020 | Change |
|---|---|---|---|
| P/E» | ~13.05x | ~14.3x | ⚠️ up - despite the lower share price, TTM EPS fell further (Rp232.12 to Rp193.31) than the price did |
| P/B» | ~2.00x | ~1.76x | ✅ down - book value per share grew while the price fell |
A same-period peer read is available: BCA's Q3 2020 post reported an annualized 9M 2020 P/E of ~25.0x against BRI's genuine TTM P/E of ~14.3x - a gap that's actually narrowed from Q2 2020's ~2.20x to roughly 1.75x on these two annualization methods (not directly comparable, but directionally suggestive). That narrowing is entirely a BRI story, not a BCA one - the clearer divergence this quarter is on coverage discipline: BCA's provision/NPL coverage rebuilt to a series-high 243.5%, while BRI's kept falling to 159.9%, yet BRI's multiple still crept up because its earnings fell faster than its price. The market's persistent premium for BCA looks, this quarter, less like a soft preference and more like a direct read on which bank is actually managing its Covid-19 credit exposure more conservatively - a premium BRI's own numbers, if anything, justify further even as the raw multiple gap narrows.
A full DCF still isn't included here, for the reason every prior post in this series has given, reinforced again this quarter: Corporate Non-SoE's NPL just posted its fourth straight series-worst reading, more than a quarter of the loan book still sits inside a regulatory-relief-adjusted risk measure whose own quality is eroding, and BRI just posted its largest provisioning quarter of the year without actually improving coverage. None of that is a stable base for multi-year credit-quality assumptions yet. The peer-multiple read above, alongside Beyond the Usual, remains the honest valuation lens this quarter.
PT Bank Rakyat Indonesia (Persero) Tbk's published financial statements as of September 30, 2020 and for the nine-month period then ended (unaudited, OJK transparency-format), together with its 9M 2020 investor presentation.