Capital One Financial Corp (COF) 2026 Q2 法說會逐字稿

Good day, and thank you for standing by. Welcome to the Capital One Q2 2026 earnings call. Please be advised that today's conference is being recorded.

各位好，感謝您耐心等候。歡迎參加 Capital One 2026 年第二季財報電話會議。敬請注意，今日會議將進行錄音。

(Operator Instructions)

(接線員指示)

I would now like to hand the conference over to your speaker today, Jeff Norris, Senior Vice President of Finance. Please go ahead.

現在我想將會議交給今天的講者，財務資深副總裁 Jeff Norris。請開始。

Thanks very much, Josh, and welcome, everyone. To access the live webcast of this call, please go to the Investors section of Capital One's website, capitalone.com. A copy of the earnings presentation, press release and financial supplement can also be found in the Investors section of Capital One's website by selecting financials and then quarterly earnings release.

非常感謝你，Josh，也歡迎各位。如欲收看本次電話會議的即時網路直播，請前往 Capital One 網站 capitalone.com 的「投資人」專區。財報簡報、新聞稿與財務補充資料亦可在 Capital One 網站的「投資人」專區中找到，請選擇「財務資訊」，再選擇「季度財報發布」。

With me this evening are Mr. Richard Fairbank, Capital One's Chairman and Chief Executive Officer; and Mr. Andrew Young Capital One's Chief Financial Officer. Rich and Andrew are going to walk you through this presentation, summarizing our second-quarter results for 2026. Please note that this presentation may contain forward-looking statements. Information regarding Capital One's financial performance and any forward-looking statements contained in today's discussion and the materials speak only as of the particular date or dates indicated in the materials.

今晚與我一同出席的有 Capital One 董事長兼執行長 Richard Fairbank 先生，以及 Capital One 財務長 Andrew Young 先生。Rich 與 Andrew 將帶領各位進行本次簡報，並摘要我們 2026 年第二季的業績。請注意，本簡報可能包含前瞻性陳述。今日討論及相關資料中所載之 Capital One 財務表現資訊與任何前瞻性陳述，僅以資料中所示之特定日期或日期為準。

Capital One does not undertake any obligation to update or revise any of this information, whether as a result of new information, future events or otherwise. Numerous factors could cause our actual results to differ materially from those described in forward-looking statements. And for more information on these factors, please see the section titled Forward-Looking Statements in the earnings release presentation and the Risk Factors section of our annual and quarterly reports accessible at Capital One's website and filed with the SEC.

Capital One 不承擔更新或修訂上述任何資訊之義務，無論係因新資訊、未來事件或其他原因。多項因素可能導致我們的實際結果與前瞻性陳述所描述者存在重大差異。如需了解更多相關因素，請參閱財報新聞稿簡報中「前瞻性陳述」一節，以及可於 Capital One 網站取得並已向美國證券交易委員會（SEC）申報之年度與季度報告中的「風險因素」一節。

Now I'll turn the call over to Mr. Young. Andrew?

現在我把電話交給 Young 先生。Andrew？

Thanks, Jeff, and good afternoon, everyone. I will start on slide 3 of tonight's presentation. In the second quarter, Capital One earned $3 billion or $4.73 per diluted common share. As a reminder, the Brex acquisition closed in early April, and we have provided additional details related to the purchase accounting in the appendix of tonight's presentation. The results for the quarter included several adjusting items related to the Discover and Brex acquisitions, which are outlined on slide 3.

謝謝你，Jeff，各位下午好。我將從今晚簡報的第 3 張投影片開始。第二季，Capital One 獲利 30 億美元，或每股稀釋普通股盈餘 4.73 美元。提醒各位，Brex 收購案已於 4 月初完成交割，我們已在今晚簡報的附錄中提供與購買會計相關的更多細節。本季業績包含數項與 Discover 與 Brex 收購案相關的調整項目，詳列於第 3 張投影片。

Net of these adjusting items, second quarter earnings per share were $5.81. Relative to the first quarter, revenue increased 4% and non-interest expense grew 7%, resulting in pre-provision earnings growth of 1%. On an adjusted basis, pre-provision earnings were flat quarter-over-quarter. Our provision for credit losses decreased $1.1 billion or 27% to $3 billion in the quarter. The provision reflects $3.7 billion of net charge-offs of $662 million.

扣除上述調整項目後，第二季每股盈餘為 5.81 美元。相較第一季，營收成長 4%，非利息費用成長 7%，使得提列備抵前盈餘成長 1%。以調整後基礎計算，提列備抵前盈餘與前一季持平。本季信用損失準備金減少 11 億美元或 27%，至 30 億美元。該準備金反映 37 億美元的淨沖銷（net charge-offs）以及 6.62 億美元。

Turning to slide 4. I'll cover the allowance in greater detail. The $662 million allowance release in the quarter brought the allowance balance to $23 billion. Our total portfolio coverage ratio decreased 26 basis points and now stands at 5.02%. I'll cover the drivers of the changes in allowance and coverage ratio by segment on slide 5.

接著看第 4 張投影片。我將更詳細說明備抵。本季釋出 6.62 億美元備抵，使備抵餘額達到 230 億美元。我們整體投資組合覆蓋率下降 26 個基點，目前為 5.02%。我將在第 5 張投影片按業務分部說明備抵與覆蓋率變動的驅動因素。

In our Domestic Card segment, we released $705 million of allowance. The coverage ratio decreased by 41 basis points and now stands at 6.99%. The decline in the coverage ratio was driven by continued favourable observed credit in the quarter and a modest decrease in the consideration given to economic uncertainties. In our Consumer Banking segment, we built $150 million of allowance. The allowance build was primarily driven by strong growth in the auto business.

在美國信用卡（Domestic Card）分部，我們釋出 7.05 億美元備抵。覆蓋率下降 41 個基點，目前為 6.99%。覆蓋率下降主要由於本季持續觀察到的信用表現良好，以及對經濟不確定性的考量略為降低。在消費金融（Consumer Banking）分部，我們增加 1.50 億美元備抵。備抵增加主要由汽車業務的強勁成長所帶動。

The coverage ratio ended the quarter at 2.39%, 3 basis points higher than the first quarter. And finally, in our Commercial Banking segment, we released $59 million of allowance. The allowance release was primarily driven by specific reserves on loans that were charged off in the quarter. The commercial banking coverage ratio decreased 8 basis points quarter-over-quarter to 1.62%.

覆蓋率於季末為 2.39%，較第一季高 3 個基點。最後，在商業金融（Commercial Banking）分部，我們釋出 5,900 萬美元備抵。備抵釋出主要由於本季已沖銷之貸款所對應的特定準備金。商業金融覆蓋率較前一季下降 8 個基點至 1.62%。

Turning to page 6. I'll now discuss liquidity. Liquidity reserves ended the second quarter at about $144 billion, down $21 billion from the prior quarter. Our ending cash position decreased by about $22 billion to approximately $55 billion. The decrease in cash was primarily driven by growth in our loan portfolio, wholesale funding maturities late in the quarter and the impacts from Brex. Our preliminary average liquidity coverage ratio was 165%, and our preliminary average net stable funding ratio was 136%.

接著看第 6 頁。我現在談流動性。第二季末流動性準備約為 1,440 億美元，較前一季減少 210 億美元。期末現金部位約減少 220 億美元至約 550 億美元。現金下降主要由於貸款投資組合成長、季末批發融資到期，以及 Brex 的影響。我們的初步平均流動性覆蓋率（LCR）為 165%，初步平均淨穩定資金比率（NSFR）為 136%。

Turning to page 7, I'll cover our net interest margin. Our second quarter net interest margin was 8.01%, 14 basis points higher than the prior quarter. The increase was largely driven by a 9 basis point impact from 1 additional day in the quarter. The remaining increase was driven by a lower rate paid on retail deposits and a $5 billion decline in average cash balances.

接著看第 7 頁，我將說明淨利差。第二季淨利差為 8.01%，較前一季高 14 個基點。此增幅主要來自本季多 1 天所帶來的 9 個基點影響。其餘增幅則來自零售存款支付利率較低，以及平均現金餘額減少 50 億美元。

Turning to slide 8. I will end by discussing our capital position. Our common equity Tier 1 capital ratio ended the quarter at 13.7%, 70 basis points lower than the first quarter. The combination of $2.7 billion of share repurchases, approximately 40 basis point impact from the Brex transaction and an increase in risk-weighted assets more than offset net income in the quarter.

接著看第 8 張投影片。最後我將談我們的資本狀況。普通股權第一類資本比率（CET1）於季末為 13.7%，較第一季低 70 個基點。27 億美元的庫藏股回購、Brex 交易約 40 個基點的影響，以及風險加權資產增加的綜合效果，均超過本季淨利的貢獻。

With that, I will turn the call over to Rich. Rich?

接下來我把電話交給 Rich。Rich？

Thanks, Andrew, and good evening, everyone. Slide 10 shows second quarter results in our credit card business. Credit Card segment results are largely a function of our domestic card results and trends, which are shown on slide 11. The Domestic Card business posted another quarter of top line growth and strong credit results. As a reminder, we closed the Discover acquisition on May 18, 2025.

謝謝你，Andrew，各位晚安。第 10 張投影片顯示我們信用卡業務的第二季結果。信用卡分部的結果主要取決於我們美國信用卡業務的表現與趨勢，詳見第 11 張投影片。美國信用卡業務本季再度實現營收成長並維持強勁的信用表現。提醒各位，我們已於 2025 年 5 月 18 日完成 Discover 收購案交割。

So period-end balances for the prior year quarter now include the addition of the Discover portfolio. for items like purchase volume and revenue, we'll still need to discuss the partial quarter impacts of adding Discover. In the second quarter, we also added Brex to the Domestic Card business and moved our small legacy Corporate Credit Card business from the commercial bank to domestic card.

因此，去年同期的期末餘額現在已包含 Discover 投資組合的加入。對於如消費金額與營收等項目，我們仍需討論納入 Discover 的部分季度影響。第二季我們也將 Brex 納入美國信用卡業務，並將我們規模較小的既有企業信用卡（Corporate Credit Card）業務自商業銀行移轉至美國信用卡。

Second quarter purchase volume grew 26% year-over-year, primarily driven by the addition of a partial quarter of discovery purchase volume. We also posted a modest acceleration in legacy Capital One purchase volume growth and benefited from modest tailwinds from the addition of Brex and the Corporate Card business.

第二季消費金額年增 26%，主要由於納入 Discover 的部分季度消費金額。我們既有 Capital One 消費金額成長也出現小幅加速，並受惠於納入 Brex 與企業卡業務所帶來的溫和順風。

Legacy Discover purchase volume grew just under 2% year-over-year. Purchase volume for legacy Capital One businesses, inclusive of adding Brex and corporate card grew about 14% year-over-year with the significant majority of the increase coming from the acceleration of underlying organic growth before the addition of Brex and Corporate Card. Ending loan balances increased 2.6% year-over-year.

既有 Discover 的消費金額年增略低於 2%。既有 Capital One 業務的消費金額（包含納入 Brex 與企業卡）年增約 14%，其中絕大部分增幅來自於在納入 Brex 與企業卡之前，基礎有機成長加速所帶動。期末貸款餘額年增 2.6%。

The legacy Discover card loans shrunk 1.5% from the prior year, in line with our expectations for the temporary brownout of discovered loan growth. Excluding Discover, ending loans grew about 5.3% year-over-year, driven predominantly by a modest acceleration in the organic growth of legacy Capital One loans and aided by the addition of Brex on Corporate Card.

既有的 Discover 卡貸款較去年同期縮減 1.5%，符合我們對 Discover 貸款成長暫時性停擺的預期。若排除 Discover，期末貸款年增約 5.3%，主要由既有 Capital One 貸款的有機成長小幅加速所帶動，並受惠於企業卡業務納入 Brex。

We continue to see good opportunities to grow the Discover Card business on the other side of our tech integration, where we can implement growth expansions powered by our unique technology and underwriting.

我們仍持續看到在技術整合完成後，Discover 卡業務有良好的成長機會；屆時我們可運用我們獨特的技術與承保能力，推動成長擴張。

Revenue was up 30% from the second quarter of 2025, largely driven by the addition of a partial quarter of Discover revenue. Excluding Discover, year-over-year revenue growth was 9.5%, driven predominantly by underlying organic growth in legacy Capital One purchase volume and loans. Revenue margin for the quarter was 17.4%. The domestic card charge-off rate for the second quarter was 4.71%, down 39 basis points from the prior quarter and down 54 basis points year-over-year.

營收較 2025 年第二季成長 30%，主要由於納入 Discover 的部分季度營收。若排除 Discover，營收年增 9.5%，主要由既有 Capital One 消費刷卡金額與貸款的基礎有機成長所帶動。本季營收利差為 17.4%。第二季國內卡片核銷率為 4.71%，較前一季下降 39 個基點，較去年同期下降 54 個基點。

The delinquency rate was 3.39% at quarter end, down 31 basis points from the linked quarter and down 21 basis points from a year ago. We are seeing similar credit trends in both the legacy Capital One and legacy Discover portfolios. Domestic Card non-interest expense was up 38% compared to the second quarter of 2025, driven by the addition of a partial quarter of Discover as well as continuing technology investments.

季末逾期率為 3.39%，較上一季下降 31 個基點，較一年前下降 21 個基點。我們在既有 Capital One 與既有 Discover 投資組合中看到相似的信用趨勢。國內卡片非利息費用較 2025 年第二季增加 38%，主要由於納入 Discover 的部分季度，以及持續的科技投資。

Operating expense and marketing both increased year-over-year. Our choices in domestic card are the biggest driver of total company marketing, but choices in our consumer banking business have an increasing impact as well.

營運費用與行銷費用皆較去年同期增加。我們在國內卡片業務的策略選擇是全公司行銷支出的最大驅動因素，但我們消費金融業務的策略選擇也正帶來愈來愈大的影響。

Total company marketing expense in the quarter was about $1.7 billion, up 23% year-over-year, driven by the addition of Discover as well as higher legacy Capital One direct marketing in our Domestic Card and Consumer Banking businesses, increased media spend and continuing investments in premium benefits.

本季全公司行銷費用約 17 億美元，年增 23%，主要由於納入 Discover，以及既有 Capital One 在國內卡片與消費金融業務的直效行銷增加、媒體投放提高，以及持續投資高端權益。

Pulling up, our marketing continues to deliver strong new account originations to build an enduring franchise with heavy spenders at the top of the domestic credit card market and to grow checking accounts on a national scale in our consumer banking business. We continue to lean into marketing to take advantage of these compelling market opportunities.

從更宏觀的角度來看，我們的行銷持續帶來強勁的新帳戶開立，以在國內信用卡市場高端、重度消費者族群中建立長期穩固的特許經營，並在消費金融業務中以全國規模擴大支票帳戶。我們將持續加大行銷投入，以把握這些具吸引力的市場機會。

Slide 12 shows second quarter results in our Consumer Banking business. Global Payment Network transaction volume for the quarter was approximately $190 billion. Network transaction volume increased 156% compared to the partial quarter of transaction volume in the second quarter of 2025. And the successful completion of Capital One debit customers -- the conversion of Capital One debit customers to the Discover network. The sequential quarter increase was about 9%. Auto originations were up 19% from the prior year quarter. We continue to be in a strong position to pursue resilient growth in the current marketplace.

投影片 12 顯示我們消費金融業務的第二季結果。本季全球支付網路交易量約 1,900 億美元。相較於 2025 年第二季的部分季度交易量，網路交易量增加 156%。以及 Capital One 簽帳金融卡客戶成功完成轉換——將 Capital One 簽帳金融卡客戶轉至 Discover 網路。季比增幅約 9%。汽車貸款承作量較去年同期增加 19%。在目前的市場環境下，我們仍處於有利位置，可追求具韌性的成長。

Consumer banking ending loan balances increased $9.2 billion or about 11% year-over-year. Average loans were also up 11%. Compared to the year ago quarter, ending consumer deposits grew about 5%. Average deposits were up 19%. Our digital-first national consumer banking business continues to grow and gain traction.

消費金融期末貸款餘額年增 92 億美元，約 11%。平均貸款亦增加 11%。與一年前同期相比，期末消費者存款約成長 5%。平均存款增加 19%。我們以數位優先、全國布局的消費金融業務持續成長並逐步取得動能。

Consumer Banking revenue for the quarter was up about 26% year-over-year, driven predominantly by the addition of a partial quarter of Discover operations as well as discover revenue synergies and growth in auto loans. Non-interest expense was up about 24% compared to the second quarter of 2025, driven largely by the addition of a partial quarter of Discover as well as higher marketing to drive growth in our National Consumer Banking business, increased auto originations and continued technology investments.

本季消費金融營收年增約 26%，主要由於納入 Discover 營運的部分季度，以及 Discover 營收協同效益與汽車貸款成長。非利息費用較 2025 年第二季增加約 24%，主要由於納入 Discover 的部分季度，以及為推動全國消費金融業務成長而提高行銷支出、汽車貸款承作量增加與持續的科技投資。

The auto charge-off rate for the quarter was 1.43%, up 18 basis points year-over-year and down 21 basis points from the sequential quarter. The year-over-year increase is the result of a gradual mix shift in new originations and loans as our subprime mix is returning to pre-pandemic levels. The auto delinquency rate was up 11 basis points from the linked quarter and down 52 basis points from the prior year.

本季汽車貸款核銷率為 1.43%，較去年同期上升 18 個基點，較上一季下降 21 個基點。年增的原因是新承作與貸款組合逐步轉變，次級貸款占比正回到疫情前水準。汽車貸款逾期率較上一季上升 11 個基點，較去年同期下降 52 個基點。

Slide 13 shows second quarter results for our Commercial Banking business. Compared to the linked quarter, both ending and average loan balances were up about 1%. Ending deposits were down about 1% from the linked quarter. Average deposits were essentially flat. The Commercial Banking net charge-off rate for the second quarter increased 24 basis points from the sequential quarter to 0.53%.

投影片 13 顯示我們商業銀行業務的第二季結果。與上一季相比，期末與平均貸款餘額皆約增加 1%。期末存款較上一季下降約 1%。平均存款基本持平。第二季商業銀行淨核銷率較上一季上升 24 個基點至 0.53%。

The commercial criticized performing loan rate was 4.4%, down 55 basis points compared to the linked quarter. The criticized nonperforming loan rate was down 8 basis points to 1.32%. In closing, second quarter results continued to reflect solid top line growth and strong credit performance. We're now 14 months into our planned 24-month integration of Discover and integration is going well with the successful completion of converting Capital One's debit customers to the Discover network. Second quarter results include the full quarterly run rate debit revenue synergies.

商業「受關注但仍正常還款」貸款比率為 4.4%，較上一季下降 55 個基點。受關注的不良貸款比率下降 8 個基點至 1.32%。總結而言，第二季結果持續反映穩健的營收成長與強勁的信用表現。我們已進入原訂 24 個月 Discover 整合計畫的第 14 個月，整合進展良好，並已成功完成將 Capital One 的簽帳金融卡客戶轉至 Discover 網路。第二季結果已包含簽帳金融卡營收協同效益的完整季度化運行水準。

Our results also include about one-third of the quarterly run rate of the announced operating expense synergies. We remain on track to deliver the full $2.5 billion of announced synergies. For years, we have been working backwards from the dramatic transformation of the business marketplace with modern technology, data, and AI. We are in the 14 year of our technology transformation from the bottom of the tech stack up. We're way down that path, and we continue to invest in some very powerful foundational capabilities as well as AI infrastructure and specific AI experiences.

我們的結果亦包含已公告營運費用協同效益之季度化運行水準約三分之一。我們仍按計畫達成已公告的 25 億美元協同效益全額。多年來，我們一直以現代科技、數據與 AI 對商業市場的劇烈轉型為出發點，倒推規劃。我們自技術堆疊底層向上推進的技術轉型已進入第 14 年。我們在這條路上已走得很深，並持續投資於非常強大的基礎能力、AI 基礎設施，以及具體的 AI 體驗。

We also continue to invest in growing our heavy spender franchise at the top of the market, including rewards, lounges, unique access to experiences and breakthrough digital capabilities. And we continue to lean into our unique quest to organically build a digital-first full-service national bank.

我們也持續投資於擴大在市場高端的重度消費者特許經營，包括獎勵、貴賓室、獨特的體驗權益，以及突破性的數位能力。同時，我們也持續推進我們獨特的目標：以有機方式打造一家數位優先、全方位服務的全國性銀行。

Many of our opportunities are enhanced by the Discover acquisition, which, of course, also brings the new opportunity to grow and scale our own global payments network. We continue to invest in network acceptance and technology. As we've discussed, these investments will continue to be reflected in the efficiency ratio.

我們的許多機會因收購 Discover 而得到強化；當然，這也帶來新的機會，讓我們得以成長並擴大自有的全球支付網路。我們持續投資於網路受理與技術。如同我們先前所討論，這些投資將持續反映在效率比率上。

They are also the engine that powers long-term growth and returns. Pulling way up, we continue to build momentum from the game-changing acquisition of Discover. Even though some individual variables in our deal model have moved since the announcement, and we have acquired Brex and brought in-house the technology that supports Capital One travel, we still expect our earnings power on the other side of the Discover integration to be consistent with what we expected at the time we announced the deal.

它們同時也是驅動長期成長與報酬的引擎。從更宏觀的角度來看，我們持續從改變遊戲規則的 Discover 收購案中累積動能。儘管自公告以來，我們交易模型中的部分個別變數有所變動，且我們已收購 Brex 並將支援 Capital One Travel 的技術內製化，我們仍預期在完成 Discover 整合後的獲利能力，將與我們在宣布交易時的預期一致。

And now we'll be happy to answer your questions. Jeff?

接下來我們很樂意回答各位的問題。Jeff？

Thanks, Rich. We'll now start the Q&A session.

謝謝，Rich。我們現在開始問答環節。

(Event Instructions) Josh, please start the Q&A.

（活動說明）Josh，請開始問答。

(Operator Instructions)

(操作員指示)

Terry Ma, Barclays.

Terry Ma，巴克萊。

I wanted to start off with Brex. Rich, you had previously indicated that you could accelerate Brex's growth almost from day one through stepped-up marketing and tech spend. So I'm just curious to what extent have those investments already been absorbed into the current expense run rate. And then when should investors see more tangible benefit become more visible. And I have a follow-up.

我想先從 Brex 談起。Rich，你先前曾表示，透過加大行銷與科技支出，你們幾乎可以從第一天起就加速 Brex 的成長。所以我想了解的是，這些投資在多大程度上已經被納入目前的費用支出水準（run rate）之中。另外，投資人何時應該會看到更具體、可見的效益。我還有一個追問。

Thank you, Terry. Just to comment on Brex for a second. I don't believe we said this from the second we get it, we we'll be able to accelerate their growth. What we said is from -- pretty much from the second that we do this acquisition, we're going to be able to start mobilizing the solutions that many of which don't require full integration and those solutions are -- can be very beneficial and help us lean in and really accelerate Brex's growth. So it's been over 100 days since we closed the deal, and we are as excited as ever about Brex.

謝謝你，Terry。先就 Brex 簡單評論一下。我不認為我們說過從我們拿到它的那一刻起，我們就能加速他們的成長。我們說的是——基本上從我們完成這筆收購的那一刻起，我們就能開始動員各項解決方案，其中許多不需要完全整合；而這些解決方案——可以非常有利，幫助我們加大投入，並真正加速 Brex 的成長。所以，距離我們完成交易已經超過 100 天，我們對 Brex 的期待一如既往地高。

We acquired Brex because of its success in the attractive corporate card market and for its bottom of the tech stack infrastructure and its world-class talent. And we just continue to be impressed with all of those striking capabilities. And together, we're making good progress in building our foundational capabilities that will support the business going forward.

我們收購 Brex，是因為它在具吸引力的企業卡市場的成功，以及它在技術堆疊底層的基礎設施與世界級人才。而我們也持續對這些突出的能力印象深刻。同時，我們正共同在建立能支撐未來業務的基礎能力方面取得良好進展。

So Brex is already experiencing some of the early tailwinds that will come with our brand, and we expect these to only grow stronger over the next few months. They are also benefiting from the cost of funds impact of moving to our balance sheet. We have already stood up a program to share high potential leads from across our businesses with and we are seeing very promising early results at the outset. We will scale into this approach more aggressively over time.

因此，Brex 已經在體驗到一些來自我們品牌的早期順風，我們預期未來幾個月這些效應只會更強。他們也受益於移轉到我們資產負債表後所帶來的資金成本影響。我們已經建立了一個計畫，將我們各項業務中高潛力的線索分享給他們，並且在一開始就看到非常令人鼓舞的早期成果。隨著時間推進，我們會更積極地擴大這種做法。

Now some other benefits that we bring are going to take a little bit longer. Over the coming months as we test and learn, we will start leaning in with marketing dollars. Fully leveraging the marketing machine of Capital One requires a little more tackle integration. We'll have to set up data pipelines and calibrate our models for Brex's customer base. So that will come a little further down the road.

現在，我們帶來的其他一些效益會需要更長一點時間。在接下來幾個月，隨著我們測試與學習，我們將開始投入行銷預算。要充分運用 Capital One 的行銷機器，需要更多的整合工作。我們必須建立資料管線，並針對 Brex 的客戶群校準我們的模型。因此，這會在稍後的階段逐步實現。

For our travel business, we will be focused on the hopper build-out through the balance of this year. So bringing our travel portal to Brex will likely follow that work. We expect that Brex will also bring many benefits to Capital One, especially bringing Brex capabilities to our small business card business. These benefits will require integration and will be unlocked over time.

就我們的旅遊業務而言，今年剩餘時間我們會專注於 Hopper 的建置擴充。因此，把我們的旅遊入口網站帶到 Brex，可能會在那項工作之後進行。我們也預期 Brex 會為 Capital One 帶來許多好處，尤其是把 Brex 的能力帶入我們的小型企業卡業務。這些效益需要整合，並會隨時間逐步釋放。

So are already moving to bring benefits to them. Most of that is not yet reflected in our investment dollars because really most of the work has been sort of working to put capabilities in place.

所以我們已經在推動把效益帶給他們。其中大部分目前尚未反映在我們的投資支出上，因為多數工作其實是在把各項能力先建置到位。

Got it. That's helpful. And then for my follow-up, regarding loan growth, that continues to improve each month in the card business even in spite of the Discover brown out. So as we kind of look ahead to Discover originations being fully on Capital One's platform, should we think about growth in the card business after that? And then also the associated marketing spend required to kickstart Discovery growth again?

了解。這很有幫助。那我的追問是，關於放款成長，即使在 Discover 出現「brown out」的情況下，卡片業務的放款成長仍在每個月改善。所以當我們展望 Discover 的核貸（originations）完全上到 Capital One 的平台之後，我們是否應該預期卡片業務的成長會如何？以及為了重新啟動 Discover 成長所需的相關行銷支出？

Thanks very much, Terry. So maybe what I'll do with your question is I think it's really getting at this thing that I proverbially call the Discover brown out. So let me just comment on that, and then I'll come back and talk about marketing spend. So as we mentioned previously, the Discover card portfolio is going through a bit of a loan growth brown out as several factors combined to pressure loan growth in the near term.

非常感謝你，Terry。所以我可能會這樣回答你的問題：我認為你真正想問的是我習慣稱之為 Discover「brown out」的這件事。我先就此做些評論，然後再回來談行銷支出。如同我們先前提到的，Discover 的信用卡組合正經歷一段放款成長的「brown out」，因為多項因素疊加，對短期放款成長造成壓力。

Following Discover's credit expansion in their card business in '22 and '23, they dialled back their origination programs and credit line management by a fair amount toward the end of 2023 and largely sustained those dial backs. Since we took over, we have been trimming on the margins of Discover's credit policy in areas where we are less comfortable with the resiliency of the underlying customers more really on the -- with respect to high balance revolvers.

在 Discover 於 2022 年與 2023 年擴張其卡片業務的授信之後，他們在 2023 年底相當程度地下調了核貸計畫與信用額度管理，並且大致維持這些下調。自我們接手以來，我們也在 Discover 的授信政策邊際上做了修剪，針對我們對底層客戶韌性較不放心的領域——更具體而言，是針對高餘額循環戶（high balance revolvers）。

As a result of these collective pullbacks, the portfolio has been contracting and continues to face some headwinds to growth as these more recent smaller vintages mature. As we mentioned Discover card out standings were down 1.5% year-over-year. Now it's worth noting that the flip side of these pullbacks and the brown out has been strong credit performance, and we're glad to see that playing through the system. So let's talk about returning to growth and getting on the other side of this brown out of Discover volume.

由於這些合計的收縮措施，該組合一直在縮減；而隨著近期較小批次（vintages）逐步成熟，成長仍持續面臨一些逆風。如我們提到的，Discover 信用卡放款餘額（outstandings）年減 1.5%。值得注意的是，這些收縮與「brown out」的另一面是強勁的信用表現，我們很高興看到這點在系統中逐步反映出來。所以我們來談談如何回到成長，並走出 Discover 量能的這段「brown out」。

The brown out is temporary since -- over time, we will bring to bear a number of capabilities as we move Discover originations and existing customers to Capital One's technology. Getting discover on to Capital One's technology will allow us to unleash our models, full spectrum underwriting and lean into spender capabilities to power more originations, higher spend volume and ultimately, higher loan volume over time.

這個「brown out」是暫時的，因為——隨著時間推進，當我們把 Discover 的核貸與既有客戶移轉到 Capital One 的技術平台時，我們將能運用多項能力。讓 Discover 上到 Capital One 的技術平台，將使我們能釋放我們的模型、全光譜承作（full spectrum underwriting），並加大投入於促進刷卡（spender）能力，以推動更多核貸、更高的消費量，並最終在時間推移下帶來更高的放款量。

And so we remain excited about the longer to potential. And let me just talk a little bit about where we are on that journey. On the Discover front book, Discover originations are now on Capital One's tech platform, and we expect to be fully on our tech stack for new originations by the end of the third quarter.

因此，我們對較長期的潛力仍然感到振奮。我也簡單談一下我們在這段旅程中的進度。在 Discover 的前端新戶（front book）方面，Discover 的核貸目前已在 Capital One 的技術平台上運行，我們預期在第三季末前，新核貸將可完全切換到我們的技術堆疊（tech stack）。

And we're now leaning into a combination of testing and rolling out capabilities that have powered our card growth at Capital One and that we believe will be enhancing to the Discover book and the Discover new flow of applicants. We are already seeing several positive green shoots, but it's early, but the -- our early read is confirmatory of our hopes there.

而我們現在正結合測試與推出一系列能力——這些能力曾推動 Capital One 的卡片成長，我們相信也會強化 Discover 的帳簿（book）以及 Discover 新申請人的流量。我們已經看到幾個正向的早期「綠芽」，但仍屬早期；不過——我們的初步判讀與我們的期待一致。

On Discover's back book, we will begin the major conversion waves later this month, but we will not be fully on Capital One's tech stack until the first quarter of next year. And so we'll have to wait a bit longer to see these benefits fully manifest. Basically, we're migrating the remainder of the back book in waves. A wave in July, a wave in October, a wave in January. So these are -- will go in phases.

在 Discover 的存量帳簿（back book）方面，我們將在本月稍晚開始主要的轉換波次，但要到明年第一季才會完全切換到 Capital One 的技術堆疊。因此，我們還需要再等一段時間，才能看到這些效益完全顯現。基本上，我們會以波次方式遷移剩餘的存量帳簿。7 月一波、10 月一波、1 月一波。所以這些——會分階段進行。

With respect to the brown out, we expect continued contraction in the near term, but as we can unleash more of Capital One's tech and capabilities with Discover on the other side of our conversions, we're looking forward to returning to growth. I do want to also mention in parallel to Discover's dial back of card loans. They also dialled back on personal loans, and we have also sort of mechanically during the integration, dial back on -- a little bit on the personal loans as well.

就「brown out」而言，我們預期短期內仍會持續收縮；但當我們在完成轉換、讓 Discover 站上我們的平台之後，能釋放更多 Capital One 的技術與能力時，我們期待能回到成長。我也想補充一點：除了 Discover 下調信用卡放款之外，他們也下調了個人貸款；而在整合期間，我們也在機械性地——在個人貸款方面也稍微下調了一些。

So that brown out will continue and, in fact, increase. And the bottom, sort of the bottom of the brown out will be somewhere around the fourth quarter of this year. But then we look forward to leaning into that growth over time. So pulling up on the brown out, they are natural and temporary part of the deal. None of them are reflective of any concerns.

因此，這個「brown out」將會持續，且事實上會加深。而「brown out」的谷底大概會落在今年第四季左右。但之後我們期待能隨時間推進加大投入、推動成長。所以總結來說，這些「brown out」是交易中自然且暫時的一部分。它們都不反映任何疑慮。

We have long term. And in fact, all of it is really just part of an integration and integrating of credit policies and look forward to stepping on the gas a little bit more gradually in the coming months. You asked Terry about marketing spend. We will lean into marketing more on the Discover side as we -- really marketing is really mostly a front book thing. So we are, as we speak, leaning more into the marketing so that we can now generate some very good flow of applicants to Capital One. So that will be one of the numerous things that we're leaning into over the course of the next year.

我們有長期規劃。而且事實上，這一切其實都只是信用政策整合與融合的一部分，我們也期待在接下來幾個月更循序漸進地稍微加大力道。你問 Terry 關於行銷支出。我們會在 Discover 端更著力於行銷——其實行銷主要是偏向前端新戶（front book）的事情。所以我們此刻正加大行銷投入，讓我們現在能為 Capital One 產生非常好的申請人流量。

Sanjay Sakhrani, KBW.

Sanjay Sakhrani，KBW。

I guess my first question is for Andrew. If I look at the NIM and sort of you alluded to this in your prepared remarks, seems like the liquidity portfolio came down over the course of the quarter ended lower, but was still high on average. So I estimate there was at least a 10 basis point plus drag on the NIM as a result. Is that a safe assumption to make? So as we enter into the next quarter, you should have a higher NIM going into the third quarter?

我想我的第一個問題是問 Andrew。如果我看淨利差（NIM），以及你在事先準備的講稿中提到的內容，看起來流動性投資組合在本季期間有所下降、期末也更低，但平均水位仍然偏高。所以我估計因此對 NIM 至少造成了 10 個基點以上的拖累。這樣的假設合理嗎？那麼進入下一季時，你們在第三季的起始 NIM 應該會更高嗎？

Thanks for the question, Sanjay. Yes, as you said, in the first quarter, we did have elevated. Cash levels from the Discover Home loan sale at the end of '25. And then we had really strong deposit growth in Q1 that was aided by tax refunds. And at that point, we ended the quarter with around $75 billion of cash. And so in the second quarter, it came down quite a bit from growth and from the maturities that I had referenced in the Q1 call as well as the cash impact related to Brex, which all of those things drove the ending balance down $20 billion, but average only came down about [5].

謝謝你的問題，Sanjay。是的，如你所說，在第一季我們確實有偏高的現金水位，來自於 25 年底 Discover 房貸出售。然後我們在第一季也有非常強勁的存款成長，並且受到退稅的助益。當時我們在季末大約有 750 億美元的現金。因此在第二季，現金水位因成長以及我在第一季電話會議中提到的到期（maturities），再加上與 Brex 相關的現金影響而大幅下降；這些因素使期末餘額下降了 200 億美元，但平均水位只下降了大約[5]。

So as you suggest, looking ahead, there should be a bit of a NIM catch-up that happens in the third quarter as the average cash catches up to the end cash. And then also, as a reminder, in the back half of the year, we have one more day in each of the quarters. So that adds a 9 basis point tailwind to NIM. And so if I just pull up on all of those things, I'd be remiss if I didn't just highlight clearly, any significant changes in our balance sheet could impact NIM over time. And our NII is almost perfectly neutral to rates over time, but if and when the Fed moves, there could be an impact to NIM at least in the short term, given the timing of the repricing of deposits and assets.

所以如你所提，往前看，第三季應該會出現一些 NIM 的追趕效應，因為平均現金水位會追上期末現金水位。另外也提醒一下，在下半年，每一季都會多一天。因此這會為 NIM 帶來 9 個基點的順風。把這些因素綜合起來，我也必須清楚強調，資產負債表若有任何重大變動，可能會隨時間影響 NIM。而我們的淨利息收入（NII）長期來看幾乎對利率變動完全中性，但如果、以及當聯準會（Fed）採取行動時，考量存款與資產重定價的時間差，短期內仍可能對 NIM 造成影響。

But that effect should level itself out over time. So last quarter, I pointed you to the back half of last year as a pretty decent proxy for a NIM level after we closed on discover. And so there will, of course, be quarterly variability from day count and other seasonal factors, but I continue to point you to that as a pretty good indicator of where our structural NIM is going to be likely in at least the near term.

但這個效果隨時間應該會趨於平衡。所以上一季我請你參考去年下半年，作為我們完成收購 Discover 後 NIM 水準的一個相當不錯的代理指標。因此，當然每季會因天數與其他季節性因素而有波動，但我仍然認為那是判斷我們結構性 NIM 在至少短期內可能落點的一個相當好的指標。

Okay. Perfect. I guess that main question from last quarter. Rich, maybe just to go back to Harry's question on expenses. I guess as we think about the incremental expenses for the investments in Brex and marketing and such.

好的。完美。我想回到上一季的那個主要問題。Rich，或許回到 Harry 關於費用的問題。我想就我們思考在 Brex、行銷等投資所帶來的增量費用而言。

Should we think about the impact to adjusted operating efficiency ratio is more marginal on a go-forward basis versus what we've seen with Brex and hopper now in the run rate? Just trying to get a sense of the margin because you do also have the remaining two-thirds of the OpEx synergies coming as we move into next year as well. So I would appreciate some color there.

我們是否應該把對調整後營運效率比率（adjusted operating efficiency ratio）的影響，視為未來更偏邊際性的影響，而不是像現在 Brex 和 Hopper 已經反映在目前的 run rate 那樣？我只是想了解一下利潤率的輪廓，因為你們也還有剩下三分之二的營運費用（OpEx）協同效益，會在明年推進時逐步實現。所以希望你能多給一些說明。

Yes. Thanks, Sanjay. The efficiency ratio will continue to reflect our revenue and expense trends, our investment imperatives, and the realization of synergies. And as we've discussed, the debit revenue synergies are essentially in the numbers. The operating expense synergies are more back loaded.

是的。謝謝，Sanjay。效率比率將持續反映我們的收入與費用趨勢、我們的投資要務，以及協同效益的實現。如同我們討論過的，簽帳金融卡（debit）收入協同效益基本上已經反映在數字裡。營運費用協同效益則更偏向後段實現。

And as we mentioned earlier, we've realized about one-third of the operating expense synergies to date, and we're on track to achieve the remaining operating expense synergies by the second half of 2027. And then, of course, we continue to lean into our investment imperative, including foundational technology, AI, and the longer-term growth opportunities created by our technology transformation, and of course, Discover and Brex.

而且如我們先前提到的，截至目前我們已實現大約三分之一的營運費用協同效益，並且我們正按計畫在 2027 年下半年前達成剩餘的營運費用協同效益。此外，當然我們也持續加碼我們的投資要務，包括基礎技術、AI，以及由技術轉型所創造的長期成長機會，當然也包括 Discover 與 Brex。

So these investments are very important to the sustained growth in terms of the company over time. So the efficiency ratio is one of many drivers of the returns of the company. With all the moving pieces, we've chosen to focus our conversation on earnings power, but -- and as we've said, we expect the earnings power of the combined company and the other side of the Discover integration to be consistent with what we expected at the time we announced the Discover acquisition, inclusive of Brex and the in-sourcing of technology that supports Capital One travel and inclusive of all these investments we've been leaning into.

因此，這些投資對公司長期的持續成長非常重要。所以效率比率只是驅動公司報酬的眾多因素之一。在這麼多變動因素之下，我們選擇把對話焦點放在獲利能力（earnings power）上；而且——如我們所說，我們預期合併後公司的獲利能力，以及完成 Discover 整合後的另一端結果，將與我們在宣布收購 Discover 時所預期的一致，並且已包含 Brex、支援 Capital One Travel 的技術內製（in-sourcing），以及我們一直在加碼的所有這些投資。

So implicit in that there needs to be an efficiency ratio that makes the numbers work. But we're not specifically guiding on that. But I think that the combined financial performance of the company continues to track with this guidance we've given on earning power come out the other side of the integration.

因此，這其中隱含著必須有一個能讓數字成立的效率比率。但我們不會針對那個比率提供具體指引。不過我認為，公司合併後的整體財務表現，仍持續符合我們所提供的、關於整合完成後獲利能力的指引。

Ryan Nash, Goldman Sachs.

Ryan Nash，高盛。

Rich, maybe to build a little bit on Sanjay's question. If you look back to when the deal was announced and you put the companies together, and you lay on synergies, it got to a return that was 20% plus or minus. Given everything that you've shared with us today, it sounds like there's some more investments that you want to make, and I understand you want to preserve optionality, but -- is the right way to think about it, this should be at least a 20% return business? And what are some of the investments that could push it higher or lower in this environment?

Rich，或許在 Sanjay 的問題上再延伸一下。如果你回頭看交易宣布時，把兩家公司合併並加入協同效益，算出來的報酬率大約是 20% 上下。基於你今天跟我們分享的所有內容，聽起來你們還想做更多投資，我也理解你們想保留選擇權，但——正確的理解方式是否是：這至少應該是一個 20% 報酬率的業務？以及在目前環境下，哪些投資可能讓它更高或更低？

So Ryan, we -- there clearly was -- Discover brings strong earnings power, and we bring a lot of synergies to this deal. So earnings power has been a very important part of the conversation and a really important part of the value equation with respect to this deal. The -- a number have -- I just want to say there are a number of variables that have moved and are moving as we go along here. The brown out on Discover loan growth, which will continue for some time, and we talked about that mitigating in coming quarters, but it is still an important factor. The flip side to loan pullbacks has been better credit performance.

Ryan，我們——很明顯地——Discover 帶來強勁的獲利能力，而我們也為這筆交易帶來大量協同效益。所以獲利能力一直是討論中非常重要的一部分，也是這筆交易價值方程式中非常重要的一環。我想說的是——有許多變數已經變動、也仍在持續變動。Discover 放款成長的「棕斷期」（brown out）將會持續一段時間；我們也談到未來幾季會有所緩解，但它仍是重要因素。而放款收縮的另一面，是更好的信用表現。

Generally, credit has been performing quite well. Capital One margins have had a strength as there's been accelerating retail deposit growth, the full Walmart P&L as part of these things. And then we've had this investment imperative, which in a sense, really as two big categories to it.

整體而言，信用表現一直相當不錯。Capital One 的利差也展現韌性，因為零售存款成長正在加速，以及 Walmart 的完整損益表（P&L）納入等因素。然後我們也有這個投資要務，從某種意義上說，它其實可以分成兩大類。

One category is really the investments in technology and AI to capture the moment to capitalize over time on the -- and the extraordinary transformation that's happening out there, and we are way down the path of our technology transformation, but there's still important investments that we are making, and we continue to lean in to that. And then on the other side, we have many of the emerging growth opportunities that are going to be a very important part of the growth and value equation over time.

其中一個類別確實是對科技與 AI 的投資，把握當下的契機，隨時間推進在——以及外部正在發生的非凡轉型中實現價值。我們的科技轉型已經走了很長一段路，但仍有一些重要投資正在進行，我們也會持續加碼投入。另一方面，我們也有許多新興的成長機會，長期來看將會是成長與價值方程式中非常重要的一部分。

So the -- but the striking thing in some ways, back to the phrase, the more things change, the more they stay the same. It is striking that out the other side of this, we expect an earnings power very consistent to what we talked about at the outset. We're not branding a precise number because there are a lot of things about Capital One performance they don't lend themselves to precise settling out with precise numbers. But when we look at the earnings power as reflected in ROTCE, we feel we're headed for a performance very consistent with what we expected along the way.

所以——但在某些方面最引人注目的事情，回到那句話：變得越多，越是一樣。很引人注目的是，走過這一切之後，我們預期的獲利能力與我們在一開始談到的非常一致。我們不會給出一個精確數字，因為 Capital One 的表現有很多面向，不太適合用精確數字來精準落定。但當我們以 ROTCE 所反映的獲利能力來看，我們覺得我們正朝著與一路以來預期非常一致的表現前進。

As part of that we are -- when I talk about the investments that we're making, and we are really leaning into that long list of investments that we talked about with an equal energy we are driving efficiency in one minus all of that across the company. And a bunch of that comes from the flip side of our tech transformation, the ability to save tech costs even as we invest in other tech costs, the savings of legacy tech costs, the efficiencies that we're driving in the operations across the business.

作為其中一部分，當我談到我們正在做的投資時——我們確實以同等的能量投入我們談過的那一長串投資，同時也在全公司推動效率提升，抵消其中的影響。其中很大一部分來自我們科技轉型的另一面：即便我們在其他科技成本上投資，也能節省科技成本——節省舊有（legacy）科技成本，以及我們在整體營運與各項業務中推動的效率。

But I just want to say that we are kind of living two lives at once here, really leaning into opportunities and really, really so carefully managing the expenses to be able to simultaneously deliver the earnings power that we expected at the outset of this deal and to be positioning ourselves to create value for our investors in the extraordinary years that are unfolding in front of us.

但我只想說，我們在這裡某種程度上是同時過著兩種生活：一方面大力把握機會；另一方面又非常、非常謹慎地管理費用，讓我們能同時交付在這筆交易一開始就預期的獲利能力，並且在未來正展開的非凡歲月裡，讓自己站在為投資人創造價值的有利位置。

Got it. Maybe as my follow-up, Rich. When I look at the capital in the slides, obviously, capital came down almost 70 basis points this quarter. But if you remove the impact of Brex, you bought back a little more stock this quarter yet, capital ratios were sort of largely unchanged. And I guess now that the deal is closed, do you think we could see a further step-up in the buyback from here? And how do you think about a path towards the slated capital targets?

了解。Rich，我接著追問一下。我看投影片裡的資本，顯然本季資本下降了將近 70 個基點。但如果扣除 Brex 的影響，你們本季回購了更多股票，然而資本比率大致上卻沒有太大變化。我想問，現在交易已經完成，你覺得接下來回購會不會再進一步加速？以及你們如何看待朝既定資本目標前進的路徑？

Yes. Ryan, I'll take that one. And let me just start by focusing on the words you ended with, which is the 11% we define as a long-term capital need as opposed to a target and we continue to think that need is 11%. We just got the recent CCAR results.

是的。Ryan，這題我來回答。我先聚焦在你最後提到的用詞：我們把 11% 定義為長期資本需求（need），而不是目標（target），我們仍然認為這個需求是 11%。我們剛拿到最新的 CCAR 結果。

But every year, when that comes out, we've just seen quite a bit of volatility looking back over the last few years that going from -- in the low 10s to 7%. And so our need is derived by our internal modelling. It's just far more stable. And as we've had for a number of years now, we continue to believe that 11% is that need, where we manage our capital at any given moment in time, factors in a variety of planning assumptions, including expectations for growth and forecasted capital accretion from earnings, regulatory environment, AOCI, stock price, the macroeconomic environment.

但每一年結果出來時，回頭看過去幾年，我們都看到相當大的波動，從——低 10% 區間到 7%。因此，我們的「需求」是由內部模型推導出來的。它穩定得多。而且如同我們過去幾年一直以來的看法，我們仍相信 11% 就是那個需求；至於在任何特定時點我們如何管理資本，會納入各種規劃假設，包括成長預期、由盈餘帶來的資本累積預測、監管環境、AOCI、股價、總體經濟環境。

But I'd also say that beyond that laundry list of specific considerations, there's also a philosophic point that we view capital as having asymmetric value, particularly in times of stress providing a ton of both offensive and defensive value in those periods.

但我也想說，除了那一長串具體考量之外，還有一個理念層面的觀點：我們認為資本具有不對稱的價值，特別是在壓力時期，能在那些期間提供大量的進攻與防禦價值。

And so this multipronged approach has enabled us to maintain a strong combination of returning capital, but also strong returns and the flexibility to take advantage of growth opportunities over time. So we are not in a race to drive it down as quickly as possible to any specific number. But hopefully, that gives you a sense of how we're thinking about capital.

因此，這種多管齊下的方法讓我們能在回饋資本的同時，也維持強勁的報酬，並保有隨時間把握成長機會的彈性。所以我們並不是在比賽看誰能把它盡快降到某個特定數字。但希望這能讓你了解我們如何思考資本。

Darrin Peller, Wolfe Research.

Darrin Peller，Wolfe Research。

Look, it looks like you included a partial quarter of Brex as well as legacy corporate card in the domestic purchase volume. So I'm just trying to triangulate if you can give us a sense what would the pro forma domestic card purchase volume growth looked like in the quarter just given the acceleration we've been seeing across the industry. I think we can calculate some of it, but a little help on some of the details would be great.

你們看起來把 Brex 的部分季度，以及既有的企業卡（legacy corporate card）都納入了美國境內的消費購買量（domestic purchase volume）。所以我想交叉推算一下：考量到我們在全產業看到的加速，你們能否讓我們了解一下，如果用備考（pro forma）口徑看，本季美國境內卡片購買量的成長大概會是什麼樣子？我想我們可以算出其中一部分，但如果能在一些細節上給點協助會很有幫助。

Yes, we didn't provide the breakdown of the specific we did provide from a purchase accounting perspective, the closing balance sheet and all the associated amortization schedules, but given the relatively small percentage of capital on to our current relatively small percentage of Brex within the context of Capital One, the P&L and balance sheet on a run rate basis, just aren't that material. And that said, we're incredibly excited about the long-term prospects of adding Brex and think that the growth that this platform provides will drive significant accretion, but we don't intend to break out any of the specifics of the P&L.

是的，我們沒有提供更細的拆分；我們確實從購買會計（purchase accounting）的角度提供了期末資產負債表以及所有相關的攤銷時程表，但鑑於在 Capital One 的脈絡下，Brex 目前占比相對小——不論是損益表或資產負債表以常態化（run rate）來看——其實都沒有那麼重大。話雖如此，我們對納入 Brex 的長期前景感到非常興奮，也認為這個平台帶來的成長將推動顯著的增益（accretion），但我們不打算把損益表的任何細項單獨拆出來揭露。

I'll just -- Darrin, let me just remind you of exactly what we said in the call, right? We did say that if you just look at the legacy Capital One domestic card business, there was a modest acceleration and that the combination of that plus the addition of Brex and corporate card was about 14% with a significant majority of that driven by the legacy piece.

我補充一下——Darrin，我提醒你我們在電話會議上到底是怎麼說的，好嗎？我們確實說過，如果你只看 Capital One 既有的美國境內卡業務（legacy Capital One domestic card business），成長有小幅加速；而把這個加速再加上 Brex 與企業卡的加入，合計大約是 14%，其中絕大多數是由既有業務所帶動。

Okay. That's helpful. Just one quick follow-up would be, I know last quarter, you had mentioned, and there were some comments earlier about expenses, but more specifically, you mentioned marketing pushed back from first quarter into the remainder of the year.

好的。這很有幫助。再一個很快的追問：我知道上季你們提到過——而且剛剛也有一些關於費用的評論——更具體地說，你們提到行銷支出從第一季延後到今年剩餘期間。

So just with this still a play in this quarter, if we could just revisit the marketing expense expected more broadly when considering the investments in Discover of Brex in recent quarters. We took the recent quarters, and we average them out given some of the timing. Is that a good way to think about run rate marketing levels for the company going forward?

所以既然本季仍在發生這種情況，我想再回頭確認一下更廣義的行銷費用預期，尤其是把近幾季對 Discover 或 Brex 的投資一併考量時。我們把最近幾季的數字拿來，因為時點因素做了平均。這樣是否是思考公司未來行銷常態水準（run rate marketing levels）的好方法？

Yes, there is seasonality in that, Darrin. And if you look back at history, no one year is perfectly the same as others, but there tends to be that upward slope and particularly in the back half of the year relative to the first half, what we were highlighting in the first quarter. was just that some of the spend that we had initially anticipated happening in the first quarter was getting pushed into the second. So we just wanted to make sure that, that point was well known. But obviously, the actual levels of marketing spend are just going to be depend on the opportunities that we see in the moment.

是的，Darrin，這裡面有季節性。如果你回頭看歷史，每一年都不會跟其他年份完全一樣，但通常會呈現上升斜率，特別是下半年相對於上半年；我們在第一季強調的，只是原本預期會在第一季發生的一些支出，被延後到第二季。所以我們只是想確保大家清楚這一點。但很顯然，實際的行銷支出水準仍會取決於我們當下看到的機會。

So I don't want to give you a perfect schedule of the percentage of the annual spend in any one quarter. But if you look back at history, there's some pretty clear trends in terms of the back half relative to the front half.

所以我不想給你一個完美的時程表，去說明全年支出在某一季會占多少百分比。但如果你回頭看歷史，下半年相對於上半年的趨勢其實相當明顯。

Richard Shane, JPMorgan.

Richard Shane，JPMorgan。

I want to pull the thread a little bit more on Brex. Rich, you talked about this pretty clearly on the call that when you -- when we think about -- and the questions have sort of lined up with this, the optimal outcome for Capital One as a stand-alone business is optimizing RTC and margin.

我想再更深入追問一下 Brex。Rich，你在電話會議上講得很清楚：當你——當我們思考——而且問題也大致圍繞在這點——對 Capital One 作為獨立公司而言，最優的結果是最佳化 RTC 與利潤率（margin）。

Brex has existed in a world for most of its life, where it was benchmarked exclusively on growth. You sort of now have to balance that within Capital One. How do you optimize the real outcome of Brex with still sort of keeping an eye on what investors really care about or seem to care about in terms of maximizing ROTCE and margin in the near term?

Brex 在其大部分生命週期中所處的世界裡，一直幾乎只以成長作為衡量基準。如今你在 Capital One 之內某種程度上必須在這之間取得平衡。你要如何在仍然關注投資人真正關心、或看似關心的—就近期期望最大化 ROTCE 與利潤率—的同時，去最佳化 Brex 的實際成果？

Well, I hope the overall objective function of Capital One isn't the maximization of ROTCE in the near term. We all have our eyes on it, and we are heading to a very good exit rate on the other side of this integration. But I want to just talk about Brex and value creation there. We all know that tech start-ups have power metrics that are not vertical earnings based. And sometimes they can feel a far, far cry from how life works in a mature public company.

嗯，我希望 Capital One 的整體目標函數不是在短期內最大化 ROTCE。我們都在關注它，而且在這次整合的另一端，我們正朝著非常好的退出水準（exit rate）前進。但我想先談談 Brex 以及那裡的價值創造。我們都知道，科技新創有一些「關鍵指標」（power metrics），並不是以垂直式盈餘為基礎。而且有時候，這些指標會讓人覺得與一家成熟上市公司運作方式相去甚遠。

But we feel that Brex's approach to creating value is very consistent with Capital One's founding approach when we created the company all the way to today. And that relates to taking a horizontal economic view. So the in the founding of Capital One, I looked at business that it's really striking that financial big banks and everything are just so focused on vertical earnings, but really banking is an annuity business.

但我們認為，Brex 的價值創造方式，與 Capital One 從創立之初一路到今天的創業方法非常一致。這與採取「水平式」的經濟觀點有關。因此在 Capital One 創立之初，我觀察到一件很顯著的事：大型銀行與整個金融業都過度聚焦在「垂直式」盈餘，但其實銀行業是一門年金型（annuity）生意。

One, invest quite a bit of money to create annuities that last for a long period of time. And so what we did was build a massive horizontal we called it horizontal accounting, basically, where as we thought about investments or originating a cohort of accounts, we estimated the lifetime economics of that, the cost to create those, et cetera. And before the investment during as it played out, and then at the end of it all, we measured it to see if indeed value is created. And this approach to rigorous financial decision-making horizontally. The investing in annuities and creating long-term value is the financial basis of how Capital One works and how we create value.

第一，要投入相當多的資金去創造能維持很長時間的年金型資產。因此我們建立了一套龐大的「水平式」—我們稱之為水平會計（horizontal accounting）—基本上，當我們思考投資或發起一批（cohort）帳戶時，我們會估算其終身經濟效益、創造成本等等。在投資之前、投資進行中、以及最後結束時，我們都會衡量，看看是否確實創造了價值。而這種以水平視角進行嚴謹財務決策的方法，投資於年金並創造長期價值，是 Capital One 運作方式與我們創造價值的財務基礎。

So when we looked at Brex, obviously, Brex -- the world was looking at their power metrics, but we rolled up our sleeves and looked at how Brex was creating valuable annuities over time that they don't have as deep and rigorous horizontal accounting system, I wouldn't expect them to. But even as recently as today, I was in a conversation talking about the continuing work we're doing on the Capital One side, looking at Brex investments and how each tranche of investment is -- looks like it's paying off over time. And our observation was these are very value creating.

所以當我們看 Brex 時，顯然外界都在看他們的關鍵指標，但我們捲起袖子深入研究 Brex 是如何隨時間創造有價值的年金型資產；他們沒有那麼深、那麼嚴謹的水平會計系統，我也不會期待他們有。但甚至就在今天，我還在一場對話中談到我們在 Capital One 這邊持續進行的工作：檢視 Brex 的投資，以及每一批（tranche）投資隨時間推進看起來如何回收、如何產生效益。而我們的觀察是：這些投資非常能創造價值。

So not every tech company's investments are value creating. But from everything we've seen, the approach Brex has especially when we onboard them to a more kind of systematic horizontal accounting system, it will fit right into the value creation philosophy of Capital One. And what we have found in building Capital One when we have these growth opportunities, is that actually, the more you really go in and measure the value creation opportunity, very often the more we invest because we can validate that these things really create value over time.

並不是每一家科技公司的投資都能創造價值。但就我們所見，Brex 的做法—尤其當我們把他們納入更系統化的水平會計體系之後—會非常契合 Capital One 的價值創造理念。而我們在打造 Capital One 的過程中發現，當出現這些成長機會時，實際上你越深入衡量價值創造的機會，往往我們就越會加大投資，因為我們能驗證這些事情確實會隨時間創造價值。

Brex is in an amazing window of opportunity. They've got a tiger by the tail. They are going after three markets at once. The commercial card market the payables marketplace and the expense management business. They're going after it with an integrated solution, strikingly that solution is something that is needed from small companies all the way to large corporations. It's an amazingly large market.

Brex 正處在一個絕佳的機會窗口。他們等於是抓住了一隻猛虎的尾巴。他們同時在攻三個市場。商業卡市場、應付帳款（payables）市場，以及費用管理（expense management）業務。他們以一體化解決方案來切入；非常顯著的是，這套解決方案從小公司到大型企業都需要。這是一個極其龐大的市場。

So we are going to lean in and provide the resources and capabilities to help Brex get even -- create even more value. But along the way, we're going to very rigorously measure to be sure that what we're investing in generates the value on the other side.

因此我們會加大投入，提供資源與能力，協助 Brex 創造更多價值。但在此過程中，我們也會非常嚴謹地衡量，以確保我們投入的資金在另一端能產生相應的價值。

But what we see continues to validate our acquisition thesis. And I want to say, too, having seen and hung around a lot of young companies over the years, I continue to be amazed at the sophistication of how this business is run. And the and the opportunity to create value here.

但我們所看到的情況持續驗證了我們的收購論點。我也想說，這些年我看過、也長期接觸過很多年輕公司，我仍然對這家公司經營管理的成熟度感到驚訝。以及在這裡創造價值的機會。

Robert Wildhack, Autonomous Research.

Robert Wildhack，Autonomous Research。

I wanted to ask about domestic card loan growth over the last several periods, just core Capital One. That's bounced around, I think, the low three's, and you said 2.6% in the second quarter. So those have all been below the longer-term trend. Can you just remind us what's behind the slowdown there? And then like bigger picture, anything structural besides law of large numbers as to why Capital One domestic card loan growth wouldn't eventually come back to the longer-term average.

我想問一下過去幾個期間的國內卡片貸款成長，就核心的 Capital One 而言。這個數字上下波動，我記得大概在 3% 出頭的低檔，你也提到第二季是 2.6%。所以這些都低於長期趨勢。你能提醒我們，放緩背後的原因是什麼嗎？然後從更宏觀的角度來看，除了「規模越大成長越難」（law of large numbers）之外，還有沒有任何結構性因素，會使得 Capital One 的國內卡片貸款成長最終無法回到長期平均水準？

So Robert, when you're talking about domestic card, you're talking overall, including Discover in our performance. So we've talked about Discover is going through a shrinking right now. So that certainly is holding back the loan growth of Capital One. The -- when we -- if I separate out the Discover brownout effect, Capital One continues to deliver very consistently solid loan growth.

Robert，你在談國內卡片時，你指的是整體表現，包含 Discover 在我們的表現之內。我們也談過，Discover 目前正處於收縮期。所以這確實在抑制 Capital One 的貸款成長。當我們—如果我把 Discover 的「降載」（brownout）效應拆開來看，Capital One 仍然持續交出非常穩健且一致的貸款成長。

It's not -- there's things when I look at the various growth metrics and compare them with the industry, looking at legacy Capital One versus a number of the leading players in the industry. On all the growth metrics, Capital One is delivering a very strong performance. There is one thing on the loan growth side that I would highlight, and it's the flip side of very good news here.

這不是—當我看各種成長指標並與同業比較，把既有的 Capital One 與產業中一些領先玩家相比。在所有成長指標上，Capital One 都展現非常強勁的表現。在貸款成長這一端，有一件事我會特別點出來，而它其實是好消息的另一面。

Payment rates have come in -- continue to come in pretty high which we always cheer for because it pays off typically in terms of stronger credit, but it does hold loan growth back a little bit. But so if we look at the metrics here, not all of which I understand we share with you, we've got -- Discover is going through a brownout and shrinking. The legacy Capital One is growing strongly on all dimensions and particularly account origination, purchase volume, a lot of the very important metrics. And then when we look -- another thing that we do, it's not something we publish, but we take the originated upmarket part of Capital One.

付款率（payment rates）一直維持在相當高的水準—而且仍然偏高；我們一向樂見其成，因為這通常會反映在更強的信用表現上，但它確實會在一定程度上抑制貸款成長。但如果我們看這裡的指標，其中有些我知道我們未必對外分享，我們看到—Discover 正在經歷降載並且收縮。既有的 Capital One 在各個面向都強勁成長，尤其是在帳戶發起（account origination）、消費金額（purchase volume）等許多非常重要的指標上。然後當我們看—我們還做了另一件事，這不是我們會公布的，但我們會把 Capital One 中「向上客群」（upmarket）發起的部分單獨拿出來看。

So we effectively proxy what the other players in the industry do who just don't go out and intentionally originate in subprime. So when we separate and look at the originated upmarket part of Capital One, this thing is absolutely humming and is right up there at the top of the league tables in the key growth metrics. So -- and that, by the way, is powered by the continued quest to win at the top of the market to win with heavy spenders. And is the flip side of our investment agenda that we have on the heavy spender side.

因此我們等於用這個方式去近似（proxy）產業中其他玩家的做法—那些玩家並不會刻意在次級（subprime）客群中發起業務。所以當我們拆分並觀察 Capital One 中向上客群的發起部分時，這一塊表現非常火熱，在關鍵成長指標上絕對名列前茅。而且順帶一提，這是由我們持續追求在市場高端勝出、在高消費者（heavy spenders）族群中勝出的策略所驅動。而這也正是我們在高消費者端投資議程的另一面。

So -- but Robert, I understand that Discover is going to hold us back for a little bit. And even on the other side, even on the other side of the integration, I think it's reasonable that that Capital One -- legacy Capital One will be a faster-growing institution than Discover. Why would that be? Just the Discover is a is a much narrower play in the credit card business is focused on the prime side of the marketplace, and it's been a really great stable play.

所以—但 Robert，我理解 Discover 會在一段時間內拖累我們。而且即便在整合完成之後、在整合的另一端，我認為合理的是：Capital One—既有的 Capital One—會是一家比 Discover 成長更快的機構。為什麼會這樣？因為 Discover 在信用卡業務上是一個更狹窄的打法，主要聚焦在市場的優質（prime）端，而且一直是一個非常出色且穩健的布局。

Capital One is legacy Capital One has got so many other growth vectors growing in card. It's probably going to continue to lead the way. But for right now, we're living with a little bit of a brown out holding our business back.

Capital One 是傳統的 Capital One，在卡片業務上還有許多其他成長動能正在發酵。它很可能會持續領先。但就目前而言，我們正承受一點「棕色停電」式的影響，拖累了我們的業務。

Don Fandetti, Wells Fargo.

Don Fandetti，富國銀行。

Rich, can you talk a little bit about the credit card migration. I know there's been some testing moving it over to the Discover network. Where are you on that? Is it encouraging? And then -- do you see a scenario where maybe you could move a little more volume over than you initially thought when you struck the Discover deal?

Rich，你能談談信用卡遷移的情況嗎。我知道你們一直在測試把它轉到 Discover 網路上。目前進度如何？有令人鼓舞嗎？另外——你是否看到一種情境：相較於當初與 Discover 達成協議時的預期，你們可能會把更多交易量轉過去？

So we're talking about Don, you're talking about moving Capital One cards to the Discover network. It always gets confusing because we're also, of course, moving Discover cards on the Capital One on platforms and things. But yes, just to clarify what we're talking about here.

所以我們在談的是，Don，你指的是把 Capital One 的卡轉到 Discover 網路上。這件事總是容易讓人混淆，因為我們當然也在把 Discover 的卡移到 Capital One 的平台等系統上。不過是的，先澄清一下我們這裡談的是什麼。

Earlier this year, we completed the conversion of our debit card business to the Discover Network, and we're very pleased with how that went. I mean I think that thing has just been I would call it a smashing success as we look at this. So now as we think about building credit card volume on the Discover network, there's two ways to do that with the front book and the back book. And -- so what we are leaning hard into right now is testing, originating legacy Capital One branded accounts on the Discover Network as well as testing the conversion of existing Capital One accounts to the Discover network.

今年稍早，我們完成了把簽帳卡（debit card）業務轉換到 Discover Network，我們對結果非常滿意。我的意思是，從我們目前的觀察來看，我會稱之為非常成功。所以當我們思考在 Discover 網路上建立信用卡交易量時，有兩種方式：新戶（front book）與存量戶（back book）。因此，我們現在正全力投入測試：在 Discover Network 上核發傳統 Capital One 品牌的新帳戶，同時也測試把既有的 Capital One 帳戶轉到 Discover 網路。

So as we lean into that and on the other side of those tests, we will then make our final choices about how -- what credit card volume that we're going to move over what timing. In parallel, an important companion of course, is scaling up the volume that the volume of investment in the network as we increase international acceptance and further build the brand.

當我們持續推進這些測試，並在看到測試結果之後，我們就會對要轉移多少信用卡交易量、以及在什麼時間點轉移，做出最終決策。同時，當然還有一個重要的配套工作：隨著我們提升國際受理度並進一步打造品牌，我們也要擴大在網路上的投資規模與投資量。

And what we're doing, the way to think about -- I mean, the quest to build international acceptance will be an always thing. So for us, the key is what we want to do is to slope the work. So we're going to slope our quest on both sides of this exercise -- with respect to acceptance, international acceptance. Well, let me, in fact, start with domestic acceptance. So Discover has -- it just blows my mind how great their domestic acceptance is.

而我們正在做的——你可以這樣理解：建立國際受理度會是一件永遠要做的事。所以對我們而言，關鍵是我們希望把工作做出「斜率」（分階段、按優先順序推進）。因此，我們會在這個專案的兩個面向都做出斜率——就受理度、國際受理度而言。其實我先從國內受理度開始說。Discover 的國內受理度之好，真的讓我非常驚訝。

There are a few scattered gaps and we are just leaning all in to literally close them all. So that's a thing that's going with great progress, and we're so pleased on the domestic side. Internationally, again, it will be a long quest. But what we're doing is sloping the -- while we're working to lift everywhere, we're particularly leaning in to lift acceptance to a higher level in the places our customers go the most. And not surprisingly, we find that where do they travel the most, they travel to Mexico, the Caribbean Canada, the UK, and those are the top four destinations. We're particularly leaning in there to really move the needle and enhance acceptance there.

只有少數零星的缺口，而我們正全力以赴，真的要把它們全部補齊。所以這件事進展非常順利，我們對國內端的成果非常滿意。至於國際方面，同樣地，這會是一段長期的旅程。但我們正在做的是把斜率拉出來——在全面提升的同時，我們特別聚焦在把受理度提升到更高水準的地區，也就是我們客戶最常去的地方。不意外地，我們發現他們最常旅行的目的地是墨西哥、加勒比海、加拿大、英國，這四個是前四大目的地。我們會特別在這些地方加大投入，真正推動顯著改善並提升受理度。

The other sloping that we're working on combined with our testing is sloping what we move and focusing more on moving things that customers or products, things that don't involve as much international travel. And so that our -- strategically, we're just working so hard to get as much volume as we can on the network. And we're going to slope the acceptance work and slope the migrations to be able to create great customer experiences and maximize the volume that we move over time.

我們正在做的另一個「斜率」安排，會結合我們的測試：在轉移內容上做出斜率，更聚焦於轉移那些客戶或產品——也就是不那麼涉及國際旅行的項目。因此，從策略上來說，我們正非常努力，盡可能把更多交易量放到這個網路上。而我們會透過對受理度工作的斜率安排，以及對遷移節奏的斜率安排，來創造良好的客戶體驗，並在時間推進中最大化我們能轉移的交易量。

And do you think you need international issuing ultimately, some suggest that you do? Or is that something you'll solve down the road?

那你認為最終是否需要國際發卡？有些人認為你們需要。還是說那會是你們之後再解決的事情？

International acceptance is there are multiple ways to build that. International issuing, by the way, is a great way to do it because -- and what we're talking about there is having a local player issue our cards. And in that way, they sort of they can help really drive the acceptance in their own local geography. So that is one of four ways to build acceptance internationally.

國際受理度有多種建立方式。順帶一提，國際發卡確實是一個很好的方式，因為——我們這裡指的是讓當地業者在當地發行我們的卡。如此一來，他們就能在自己的本地市場中，協助真正推動受理度。所以，這是提升國際受理度的四種方式之一。

In fact, again, I'm amazed at how Discover with their relatively small scale built the impressive international (technical difficulty), but it's still not yet where we would love it to be as a destination. So the ways to get there from here and Discover has used all four of these. One is partnering with other networks. And this has been a really important part of Discover's strategy. They have partnered with networks in Japan, China, India, I mean there is massive acceptance in some of the biggest countries in the world coming from network partnerships.

事實上，我再次感到驚訝的是，Discover 以相對較小的規模，建立了令人印象深刻的國際（技術問題），但距離我們希望它成為的目的地水準仍有差距。所以從現在到達那個目標的路徑，而 Discover 其實四種方式都用過。第一種是與其他網路合作。這一直是 Discover 策略中非常重要的一部分。他們與日本、中國、印度等地的網路合作——也就是透過網路合作，在全球一些最大的國家取得了非常廣泛的受理度。

A second way to do it is with card-issuing financial institutions, American Express has particularly leaned into this approach. It's a great approach. We have some cases of that with Discover. It's been less of a lever for Discover than for MX, but that's another one. And by the way, just a small point as an issuer ourselves in Canada and in the UK.

第二種方式是與發卡金融機構合作；美國運通（American Express）特別大力採用這種做法。這是一個很好的做法。Discover 也有一些這樣的案例。這對 Discover 來說不像對 MX 那樣是一個主要槓桿，但這是另一種方式。另外補充一點：我們自己也在加拿大與英國做為發卡行。

We look forward to getting our own issuer boost there on acceptance. A third lever is partnering with Mergent acquirers. And finally, the fourth is going directly to merchants. So this is the playbook Discover is used. We will continue to invest in this lay book. And there's, of course, a flywheel benefit that comes with the more acceptance we get, the more volume we can get to how that flywheel works. But those will be the four levers that we lean into in this journey.

我們也期待在那裡透過我們自身的發卡行身分，帶動受理度的提升。第三個槓桿是與商戶收單機構（merchant acquirers）合作。最後第四個是直接與商戶合作。這就是 Discover 使用過的作戰手冊。我們會持續投資這套作戰手冊。而且，受理度越高，我們能帶來的交易量就越多，這之間當然會有一個飛輪效應，推動飛輪持續運轉。但在這段旅程中，我們會主要倚重並加大投入的，就是這四個槓桿。

John Pancari, Evercore.

John Pancari，Evercore。

Back to the investments that you're making. I understand you're unable to provide for the efficiency ratio or expense growth expectations regarding your investments into the network. But any way you can help us with what inning you're in, in terms of the investments? I know, Rich, you've said in the past that these would be sustained investments for a number of years. Any additional color on where you stand now, now that you've been down the path, you've seen debit migration.

回到你們正在進行的投資。我理解你們無法提供效率比率或費用成長的預期，來說明你們對網路的投資。但你們能否用任何方式協助我們理解：就投資進程而言，你們現在處於第幾局（inning）？我知道 Rich 你過去曾說過，這些會是持續多年的投資。在你們走過這段路、也看到簽帳卡遷移之後，能否再多提供一些現況的說明？

You've talked about the testing now, and you've just walked us through in that previous answer of some of the approaches. What inning are you in with how you look at the investment required here?

你們談到目前正在做測試，而你剛剛在上一個回答也帶我們走過一些做法。就你們看待這裡所需投資而言，你們現在處於第幾局？

Well, the first thing I want to say is, when I give the big list of investments, and I know for the last sort of our whole lives at Capital One. We've always, in some ways, has been the company that's investing in our future, but there's certainly been a lot discussion, as you all have noticed about the long list of investments that we're leaning into.

嗯，我首先想說的是，當我列出那一長串投資項目時——而且我知道，在 Capital One 的這些年裡，我們在某種程度上一直都是那家投資未來的公司——但你們也注意到了，大家確實一直在討論我們正在加碼投入的那份很長的投資清單。

The first thing I want to say is, I wouldn't want anyone to draw the perception that like massively moving the needle of Capital One investments is investing in the network or international acceptance. It is an important sustained investment we will do for as far out as we can see. But I wouldn't want to leave the impression that's like at the top of the list of what we -- we're spending a lot more money than that on Capital One technology, AI and maybe the biggest single item is -- well, I don't know, there's several, but investing to win with heavy spenders at the top of the market.

我首先想說的是，我不希望任何人形成一種印象：好像大幅改變 Capital One 投資力度的關鍵，是投資於網路或國際受理度。這確實是一項重要且我們在可預見的未來都會持續進行的投資。但我不希望留下的印象是：它好像是我們清單中最上面的那一項——我們在 Capital One 的科技、AI 上花的錢要多得多；而且也許最大的單一項目是——嗯，我不確定，可能有好幾項——但其中之一是為了在市場高端、重度消費者（heavy spenders）族群中取勝而進行的投資。

So I want to say this is just one of the many things on the list. That said, the -- to your point, I believe that the -- for as far out as we can see, we'll be investing in international acceptance. But here's the key thing. We're not -- our strategy is not hinging on we have to invest so much to get to a point where then finally, we can have this big bang moment and move a whole bunch of customers. This is why I went back to the power of the sloping.

所以我想說，這只是清單上眾多事項之一。話雖如此——就你提到的點而言，我相信——在我們可預見的未來，我們都會持續投資於國際受理。但關鍵在這裡。我們不會——我們的策略並不是押注於必須投入大量資金，先到達某個點，然後終於迎來一個「大爆發」時刻，把一大批客戶一次性移轉過去。這也是為什麼我回到「斜率」的力量。

We take our customers and cards and just analyse what customers are international travellers, we can empirically see that. Some customers have never travelled outside of the country for 20 years. I mean we can see and really understand where they're coming from. We have good ways to understand on the front book what is happening. And it's partly a customer point and a product that they're choosing point.

我們拿我們的客戶與卡片來分析，哪些客戶是國際旅客，我們可以用數據實證看得很清楚。有些客戶在過去20年從未出國旅行。我的意思是，我們看得到並且能真正理解他們的來源與行為。我們有很好的方法去理解在前端新發卡組合（front book）正在發生什麼。而這部分既與客戶本身有關，也與他們所選擇的產品有關。

And then -- when we look at where customers travel, that also is so sloped. So again, I think that while we will be investing as far out as we can see in the network, by sloping the investment, we can get a lot of progress in a focused way and continue to create the ability to move more customers as soon as we can.

然後——當我們看客戶去哪裡旅行時，那同樣呈現非常明顯的「斜率」。所以再一次，我認為，雖然我們會在可預見的未來持續投資於網路，但透過把投資做成「斜率化」，我們可以用更聚焦的方式取得很大進展，並且持續創造能力，讓我們能在最早可行的時間點移轉更多客戶。

And in that way, we don't have to wait for some day to try to monetize the power of this network. We're already living it on the debit side, and we can lean into it on the credit card side and the benefits accrue right along the way with the investments.

如此一來，我們不必等到某一天才去嘗試把這個網路的力量變現。我們在簽帳金融卡（debit）端已經在實際運用它，而在信用卡端我們也可以加大力道，並且效益會隨著投資一路累積。

Okay. And then just separately, regarding the migration comments that you answered in the previous question. What of the back book cards that you would ultimately move over, how would you approach the testing and then ultimately moving over migrating capital on back book over to the Discover network. Would it start with the basic non-premium cards? And would you only focus on those that are expiring in a given year, and that's how you would focus on the migration of that back book eventually?

好的。另外，關於你在上一題回答的移轉（migration）相關評論。對於你們最終會移轉過去的存量卡（back book），你們會如何進行測試，並最終把存量資產從存量卡組合移轉到 Discover 網路？會先從基本的非高端卡開始嗎？你們是否只會聚焦在某一年到期的卡，並以此作為最終移轉該存量卡組合的方式？

So well, that's a very astute question that you asked there. So let's talk about this. First of all, like we always do in testing, we do a lot of testing just so that we understand what customer reactions to various things are going to be. So our testing is pretty broad. So that we can then not find out later, we were too narrow because we didn't think expansively enough from a testing point of view.

嗯，你這個問題問得非常精準。我們來談談這個。首先，就像我們一貫的測試方式一樣，我們會做大量測試，確保我們理解客戶對各種變動可能會有什麼反應。所以我們的測試範圍相當廣。這樣我們就不會在之後才發現，因為從測試角度看得不夠全面而導致範圍過於狹窄。

Then if you look at the factors to consider in migration. So you -- first of all, the front book is -- it's a much easier thing -- well, it's a much more straightforward thing to talk about the front book because we can just put certain cards, certain customers on the Discover network, and there isn't a migration event. And so that's a very attractive way to build business.

接著，如果你看移轉時需要考量的因素。所以你——首先，前端新發卡組合（front book）——這是一件更容易——嗯，更直觀的事情，因為我們可以直接把某些卡、某些客戶放到 Discover 網路上，而不需要發生一個「移轉事件」。因此，這是一種非常有吸引力的方式來建立業務。

When we're talking about migrating the existing book, which also is attractive, the key levers there -- the key factors to consider is international travel, that's at the top of the list, how extensively our cards are on file because the more cards that the customer has on file, the more friction there is in changing card numbers and a natural -- and one thing we're looking at, in some cases, is moving at exploration time because one would already some of those frictional elements would already be there at that time. So all of these things are part of our test agenda and our strategic considerations.

當我們談到移轉既有存量組合（existing book）時——這同樣很有吸引力——其中的關鍵槓桿、關鍵考量因素包括：國際旅行，這是清單最上面的一項；以及我們的卡片被客戶「綁定/留存於檔案」（on file）的程度，因為客戶留存於檔案的卡越多，更換卡號的摩擦就越大；另外——我們正在看的其中一件事是，在某些情況下，會在到期（expiration）時點進行移轉，因為在那個時間點，部分摩擦因素本來就會存在。所以，所有這些都納入我們的測試議程與策略考量。

Mihir Bhatia, Bank of America.

Mihir Bhatia，美國銀行（Bank of America）。

I wanted to ask about -- touch on credit for a second. And I'll just ask both the parts of my question upfront. Just firstly, on the June loss rate, it was down quite a bit, month-over-month, I think by 25 basis points. Anything to call out there? Was there a sale or something, or was that just how much better credit got there?

我想問一下——先談談信用表現。我先把問題的兩個部分一起問。第一，6月的損失率（loss rate）月比下降不少，我記得大概是25個基點。這裡有什麼需要特別說明的嗎？是有出售（sale）之類的因素，還是單純信用表現改善得那麼多？

And then just the second part was just pulling up, Rich, if you could just talk about how the consumer is caring, but more importantly, how the Capital One customer is faring? Like you've been investing a lot in marketing, growing it, are recent vintages performing in line to what you expected? Just any comments on that.

第二部分是想請 Rich 你談談消費者的狀況，但更重要的是 Capital One 客戶的狀況如何？你們在行銷上投入很多、也在擴張，近期批次（recent vintages）的表現是否符合你們的預期？有任何評論都可以。

So Mihir, let's start with the June performance. I don't have the June loss rate number right in front of me. But here is a comment about the quarter and about June. So obviously, credit continues to come in very strong. Probably the single indicator we look at the most is delinquencies. And in our card business, the -- well, the June loss rate was particularly strikingly strong.

所以 Mihir，我們先從6月的表現談起。我手邊沒有6月損失率的數字。但我可以就本季以及6月做一些評論。很明顯，信用表現持續非常強勁。我們可能最常看的單一指標是逾期（delinquencies）。而在我們的卡業務中——嗯——6月的損失率特別強勁、非常醒目。

The June delinquencies for the month, moved in line with seasonality. And by the way, in pretty much every month prior over the course of 2026. The delinquencies have moved a little bit better than our calculated seasonality. So June, again, a very strong month, but I just want to point out it's the first month that didn't actually beat seasonality.

6月當月的逾期表現，符合季節性走勢。順帶一提，在2026年迄今之前幾乎每一個月份，逾期表現都比我們計算的季節性略好一些。所以6月同樣是非常強的一個月，但我想指出，這是第一個沒有實際優於季節性的月份。所以6月同樣是非常強的一個月，但我只是想指出，這是第一個沒有實際跑贏季節性的月份。

But it's still -- there's great strength there and the charge-offs were amazing and all of that. So again, we just see a very positive credit picture, but I just wanted to make those comments about June.

但即便如此——整體仍然非常強勁，核銷（charge-offs）也非常出色等等。所以再一次，我們看到的是非常正面的信用圖像，只是我想就6月補充這些說明。

And Rich, I could just interject, there's nothing to call out in the June domestic card charge-off rate.

Rich，我插一句，6月國內卡（domestic card）核銷率沒有任何需要特別說明的事項。

Okay. Yes. So the -- let's talk about the consumer and then let's turn to Capital One customers. So the US consumer and the overall economy remained resilient despite the high energy prices and everything when you pick up the news every day, one would think the world is falling apart.

好的。是的。所以——我們先談消費者，然後再談 Capital One 客戶。美國消費者與整體經濟在高能源價格等因素下仍然具韌性；每天看新聞會讓人覺得世界快要崩潰了。

But actually, the portfolio -- that the consumer continues to perform remarkably well. The unemployment rate in June was lower than in February before the year-round conflict began. Jobless claims remain low. Job creation has rebounded over the past few months. Consumer spending, that remains strong. Now as a result of inflation, real wage growth turned negative in April and May on a year-over-year basis, but it was back in positive territory ever so slightly in June as inflation ticked back down.

但其實，投組——也就是消費者的表現仍然好得驚人。6月失業率比全年衝突開始前的2月還要低。初領失業金申請仍然偏低。過去幾個月的就業創造已經反彈。消費支出仍然強勁。不過，由於通膨，實質薪資成長在4月與5月以年增率來看轉為負值，但隨著通膨在6月再度回落，實質薪資成長也略微回到正值區間。

When we look at bank balances and debt servicing burdens of our customers, these look a bit stronger than a year ago across income levels. In our Domestic Card business, our credit metrics continued to improve on a year-over-year basis in the quarter. And the -- I've chatted a little bit about that, but the strong credit performance also -- the strong credit performance we also saw on the auto side. Auto credit metrics are strong as well.

當我們看客戶的銀行存款餘額與債務償付負擔時，跨各收入層級來看，這些指標都比一年前略強。在我們的國內卡業務中，本季信用指標在年比基礎上持續改善。而且——我剛剛也稍微談到——我們在汽車端也看到同樣強勁的信用表現。汽車信用指標也很強。

And so what I want to do now is turn to leading indicators when we look at our own customers. So we talked about delinquencies. We talked about how strong delinquency performance has been pretty much every quarter this year. Other metrics that we look at, payment rates, I talked about that earlier. They are meaningfully above pre-pandemic levels across all of our customer segments, and that is a -- slows down growth a little bit, but it's a healthy sign of customer credit quality.

所以我現在想做的是，當我們觀察自家客戶時，轉向一些領先指標。我們談到了逾期。我們談到今年幾乎每一季逾期表現都相當強勁。我們還會看的其他指標包括付款率，我先前也提過。在我們所有客戶分群中，付款率都明顯高於疫情前水準；這會讓成長稍微放慢一些，但這是客戶信用品質健康的訊號。

Spend levels. We continue to see healthy fee spend growth driven both by account growth and by steady growth in spend per customer. When we look at revolve rates, revolve rates have stabilized over the past year at close to pre-pandemic levels for our major products and segments. Now none of these observations are conclusive on their own, but I think collectively, they paint a picture of strength of the consumer and certainly strength within our own portfolio.

支出水準。我們持續看到健康的手續費型支出成長，這同時受惠於帳戶成長以及每位客戶支出穩定增加。當我們看循環信用（revolve）比率時，過去一年在我們主要產品與客群中，循環比率已穩定在接近疫情前水準。這些觀察單獨來看都不是定論，但我認為合在一起，它們描繪出消費者的韌性，以及當然也顯示我們自家投組的強勁。

Let me turn now to the front book of new originations in our Card business. Our front book of new originations continues to perform strikingly well. We're seeing our '24 and '25 originations, frankly, in both legacy Capital One and Discover.

接下來我談談我們信用卡業務中新發放（新承作）的前端帳簿（front book）。我們的新承作前端帳簿持續表現得非常出色。坦白說，我們看到在傳統 Capital One 與 Discover 兩邊，2024 年與 2025 年的承作表現都很不錯。

Well, let me separate it out. In legacy Capital One, we're seeing our 2024 and 2025 originations coming in better than 2022 and '23 and a bit below pre-pandemic levels, which is pretty striking given -- and that is not a thing that I think is being universally observed in the Card business, but it's a thing that we have seen strength in our originations for really throughout this whole post pandemic period. And it's one of the things that gives us the confidence to lean into our originations, spend that money on marketing that we talked about, et cetera.

我先把它拆開來說。在傳統 Capital One 這邊，我們看到 2024 年與 2025 年的承作表現優於 2022 年與 2023 年，且略低於疫情前水準；考量到——這點相當引人注目。而且我認為這並不是信用卡業務中普遍都能觀察到的現象，但我們確實在整個後疫情期間都看到承作的強勁。這也是讓我們有信心加大承作力道、投入我們提到的行銷費用等等的原因之一。

One of the (technical difficulty) from a credit is recoveries. Coming out of the pandemic, our recoveries inventory was unusually low, but that inventory has increased rapidly over the past couple of years and has contributed to the improvement in our overall loss rate over that time. Discover's losses peaked later than legacy Capital Ones, and they're now seeing the same dynamic be a tailwind to their losses. But looking ahead, our recoveries inventory should taper off a bit from -- in the next year or so because the inventory of recent charge-offs will itself be going down. So that's -- Mihir, that's a look at leading indicators.

（技術問題）從信用角度來看，其中一項是回收（recoveries）。疫情結束後，我們的回收存量（recoveries inventory）異常偏低，但過去幾年該存量快速增加，並在這段期間推動我們整體損失率的改善。Discover 的損失高峰出現得比傳統 Capital One 更晚，而他們現在也看到同樣的動態，對其損失形成順風。但往前看，未來一年左右我們的回收存量應會稍微趨緩，因為近期核銷（charge-offs）的存量本身將會下降。所以——Mihir，以上是對領先指標的說明。

But if I pull way up, we see real strength in the consumer and strength across our business performance in card and in auto. And that's why while we keep a very wary eye on the economy and an international development, we are leaning in with a lot of positivity into our growth strategies.

但如果我把視角拉高來看，我們看到消費者確實很強勁，而我們在信用卡與汽車業務的經營表現也同樣強勁。因此，雖然我們仍會非常謹慎地關注經濟與國際情勢發展，但我們會以相當正向的態度加大推進我們的成長策略。

Erika Najarian, UBS.

Erika Najarian，UBS。

I wouldn't prolong this call if this question wasn't important. But I think investors really want clarity on this. And Rich and Andrew, you keep mentioning that your earnings power is expected to be the same as you anticipated when you first announced a Discover deal. So I was hoping to unpack that a bit. So I was looking through your disclosures, I wasn't sure what you were using for the baseline.

如果這個問題不重要，我也不會延長這通電話會議。但我認為投資人真的希望在這點上得到更清楚的說明。Rich 和 Andrew，你們一直提到，你們預期的獲利能力與最初宣布收購 Discover 時的預期相同。所以我希望能把這點拆解一下。我在看你們的揭露資料時，不太確定你們用的是哪個基準。

But in 2027, consensus EPS -- sorry, at the deal announcement, consensus EPS for Capital One stand-alone is about $21. You mentioned over 15% accretion to [27] EPS at the announcement that rounds up to, let's call it, like [$24.50] if we use 16%, 17% accretion. Last quarter, you mentioned that when you were thinking of ROTCE, you weren't thinking of CET1 all the way down to 11%, so if you use 12.5% on the current share count, you can get to a mid-20s ROTCE pro forma. What is wrong with that line of logic?

但在 2027 年，共識 EPS——抱歉，在交易宣布時，市場對 Capital One 單獨（stand-alone）的共識 EPS 大約是 21 美元。你們在宣布時提到對 [27] 年 EPS 有超過 15% 的增益（accretion），如果用 16%、17% 的增益來算，四捨五入大概就是我們說的約 [24.50 美元]。上季你們提到，在思考 ROTCE 時，你們並不是假設 CET1 一路降到 11%；所以如果用 12.5% 並以目前股數計算，可以得到一個備考（pro forma）的 20% 中段（mid-20s）ROTCE。這樣的推論哪裡有問題？

Well, Erika, let me just unpack a couple of the assumptions that you made that don't actually tie to things that we've said, and I just want to be really clear here. First of all, our assumptions that we laid out, and I'd encourage you to go back when we announced the deal in February of '24. But what we said was we were taking consensus estimates for both Capital One and Discover, and we made the adjustment to Discover's loss forecast just based on the things that we had seen during diligence.

Erika，我先拆解一下你提出的幾個假設，這些其實並不完全對應到我們說過的內容，我想在這裡講得非常清楚。首先，我們當時提出的假設——我也建議你回去看我們在 2024 年 2 月宣布交易時的說法。我們當時說的是，我們採用 Capital One 與 Discover 兩者的市場共識預估，並且僅基於我們在盡職調查期間看到的情況，對 Discover 的損失預測做了調整。

Then with respect to ROTCE at the time, the weighted average consensus for CET1 was 12.5%. And so when Rich last quarter highlighted that we're defining earnings power as ROTCE, we just wanted to remain consistent with that nominator of 12.5% for the sake of doing the math that is not saying that that is our target.

接著就當時的 ROTCE 而言，市場共識對 CET1 的加權平均是 12.5%。因此，上季 Rich 強調我們將獲利能力（earnings power）定義為 ROTCE 時，我們只是為了計算一致性而沿用 12.5% 這個分母來做數學推導，並不是說那就是我們的目標。

As I answered before, in terms of what we believe our capital need is, we believe our capital need is 11%, but we are just doing the math on ROTCE at 12.5% for the sake of comparability. And so with respect to EPS, of course, share price assumptions have moved. There's just a number of things that have moved in terms of those assumptions, particularly as they relate to individual line items, and that is why we keep coming back to like the ROTCE as our definition of earnings power.

如我先前回答的，就我們認為所需資本水準而言，我們認為需要的是 11%，但為了可比性，我們在 ROTCE 的計算上仍以 12.5% 來做數學推導。至於 EPS，當然，股價假設已經變動。還有許多假設也都變動了，特別是與各個損益項目相關的假設；這也是為什麼我們一直回到以 ROTCE 作為我們對獲利能力的定義。

Moshe Orenbuch, TD Cowen.

Moshe Orenbuch，TD Cowen。

Rich, you talked about growth in the non-prime oil and in the high-end card business. Could you talk a little bit about the non-prime card business because that's been a business when you've grown it. It hasn't required as much upfront investment as the high end, just -- and you've also talked about the consumer doing relatively well. So are there prospects for acceleration there? And I've got a follow-up.

Rich，你談到在非優質（non-prime）以及高端信用卡業務的成長。你能否多談一點非優質信用卡業務？因為那一直是你們在擴張時，相較高端業務不需要那麼多前期投資的業務——另外你們也談到消費者狀況相對不錯。

Moshe, thank you. I know that you are on over the many decades, we have worked together that has such a such an interest in this, and it's a very, very appropriate interest because it is a very important part of Capital One, even though you saw that the percentage is down to what 26% did we see it because it dropped down a bit because of the Discover portfolio. Just in terms of the portfolio, subprime percentage.

Moshe，謝謝你。我知道在過去幾十年我們合作的過程中，你一直對這塊非常關注，而且這是非常、非常恰當的關注，因為它是 Capital One 非常重要的一部分；即便你看到占比降到大概 26%——我們看到它下降了一些，是因為納入了 Discover 的投組。就投組而言，次級（subprime）占比。

But Moshe, across -- well, in both card and auto, we continue to our strategy has remained very much the same. We continue to get a lot of traction in the business. Performance continues to be strong. When I point at the higher growth at the top of the market, that is really just pointing out the traction that Capital One is getting in our investments at that part of the market, but we continue to be very pleased with how things are going at the lower end of the market.

但 Moshe，整體來說——在信用卡與汽車兩個業務上，我們的策略基本上維持不變。我們持續在這項業務上獲得很大的進展。表現仍然強勁。當我提到在市場高端的成長較快時，那只是指出 Capital One 在該市場區段的投資正在取得成效；但我們也持續對在市場較低端的進展感到非常滿意。

The growth rates are a little lower these days than -- they're lower than what we see at the higher end of the market, but we always give -- take what the market has to give us. But performance is stable. Credit performance in that part of the marketplace, I should have mentioned this earlier, is very consistent really across the credit spectrum. We don't, in our own numbers, see this K-shaped economy that a lot of people talk about, although to be fair, we don't really participate in the lowest end of the marketplace where maybe those things are being experienced in the economy.

目前的成長率比起——比我們在市場高端看到的要低一些。但我們一向是市場給多少就拿多少。不過表現是穩定的。我應該更早提到，該市場區段的信用表現其實在整個信用光譜上都非常一致。我們從自家數據並沒有看到很多人所說的那種 K 型經濟；不過公平地說，我們並未真正參與市場最底端的區段，而那裡可能確實正在經歷這些經濟現象。

So Moshe, things continue to go very well. We're leaning in the marketing efficiency of that part of the business is, it's a lot less costly to acquire accounts there, and we continue to lean in really very hard there. And the growth is very solid. It's a little less than at the high end.

所以，Moshe，情況持續進展得非常順利。我們正加大投入，因為這部分業務的行銷效率很高——在那裡獲取帳戶的成本低得多，而我們也持續非常積極地加碼投入。而且成長非常穩健。只是比高端市場略低一些。

It's less than at the high end, but the value creation continues to be high and everything about it seems quite stable. And also one other thing this part of the marketplace is so benefited by continued investments in technology, data and the power of machine learning and overtime AI because this is all this part of the marketplace is all about data, analytics, modelling, and that is a power alley of Capital One.

確實比高端市場低一些，但價值創造仍然很高，而且各方面看起來都相當穩定。另外還有一點：這一市場區隔非常受益於我們持續在科技、數據，以及機器學習乃至長期的 AI 上的投資，因為這個市場本質上就是圍繞數據、分析與建模，而這正是 Capital One 的強項領域。

So while you -- the marketing investment isn't maybe the highest in that part of the marketplace, there's a lot of focus in our tech and data and AI investments to be able to be even more successful in that underserved part of the market. Thank you for your question.

因此，雖然在這個市場區隔的行銷投入未必是最高的，但我們在科技、數據與 AI 投資上投入了大量心力，以便在這個服務不足的市場中取得更大的成功。謝謝你的提問。

Sure. And maybe just as a quick follow-up. You talked earlier about the horizontal P&Ls that you kind of do for each of your products. And when you think about how Capital One as a company is viewed externally, I mean do you think you get recognition for the streams of earnings that you're creating and the value that, that's creating. And if not, would there be a way whether it's some degree of disclosure of examples of that?

當然。也許再快速追問一下。你先前談到你們會針對每個產品做某種橫向的損益表（P&L）。當你思考外界如何看待 Capital One 這家公司時，我的意思是，你覺得市場是否有認可你們正在創造的獲利來源，以及其所創造的價值？如果沒有，是否有什麼方式——例如某種程度的揭露、或提供一些例子——來呈現這些？

Like are there -- is there -- I mean, do you think that you're getting appropriate recognition in the stock for it? And what could you do about it?

例如——我的意思是——你覺得股價是否對此給予了適當的認可？你們能做些什麼？

Moshe, it's a great question. I believe that we probably don't get appropriate recognition in the stock. But I think the -- and I think it would -- I don't think there's an easy way for us to sort of publish the aspects of our horizontal accounting, but I would say this.

Moshe，這是個很好的問題。我相信我們的股價可能確實沒有得到適當的認可。但我認為——而且我也不覺得有一個簡單的方法，能讓我們把橫向會計的各個面向「公開發表」出來；不過我想說的是這個。

When you look at the -- I guess, we have -- I'm in my -- what is it, 32 year of running this company. Well, since we had our IPO in 40 year overall in building this franchise and one of the very, very first things we did was put in horizontal accounting and an NPV-based methodology for everything that we do.

當你回頭看——我想，我們——我現在是——算起來是經營這家公司第 32 年了。嗯，自從我們 IPO 以來；而整體打造這個特許經營（franchise）已經 40 年了。我們最早、最早做的事情之一，就是導入橫向會計，以及以 NPV（淨現值）為基礎的方法論，來衡量我們所做的每一件事。

And I think it's hard to prove that -- the power of that to investors, but I think maybe the power manifest in the 3.5-decade history of Capital One and the ability to grow the company so significantly and to generate strong earnings power along the way.

我認為要向投資人證明這套方法的威力並不容易，但我想它的威力或許已經體現在 Capital One 35 年的歷史中：公司能夠大幅成長，並在過程中持續產生強勁的獲利能力。

And the cornerstones of that approach have been starting with strategy, making sure that the businesses that we're in lend themselves to above -- they are structurally attractive and give the opportunity to generate above-hurdle returns, which is why we don't do half the things other banks do.

而這種做法的基石，是從策略出發：確保我們所進入的業務具備結構性的吸引力，並提供創造超過門檻報酬（above-hurdle returns）的機會——這也是為什麼我們不做其他銀行做的一半事情。

And then secondly, the whole investment philosophy that we have that the strategic philosophy that we focus on long-term value. The financial horizontal P&L investment approach that we use. And the way that over time, we have a whole methodology of retrospective measurement of how our various programs are performing relative to expectation, relative to hurdle rate and all of these kind of things in a way that I think has really demonstrated the power of this. And so when I then say to investors, we are at a time where we have exceptional opportunity going forward.

第二，是我們的整體投資哲學與策略哲學：我們聚焦於長期價值。我們所採用的財務橫向損益表（P&L）投資方法。以及隨著時間推進，我們有一整套回溯衡量的方法論，用來檢視各項計畫相對於預期、相對於門檻報酬率（hurdle rate）等的表現；我認為這些都真正展現了這套方法的威力。因此，當我對投資人說，我們正處於一個未來擁有非凡機會的時點。

That opportunity, there's sort of two different buckets of investment related to what I'm describing as just extraordinary opportunity we see going forward. One is on the business side, the horizontal P&L measurement of our investments across these emerging businesses and so on. And the other thing is a choice that we do at Capital One, which doesn't lend itself to such precise horizontal P&L, Moshe, as you know, which is building the technology foundation of the company.

這個機會，大致可分成兩個投資「桶」（兩大類），與我所描述的、我們所看到的未來極其非凡的機會相關。一個是在業務面：用橫向損益表（P&L）去衡量我們在這些新興業務等領域的投資。另一個則是 Capital One 的一項選擇——如你所知，Moshe——它不太能用如此精準的橫向損益表來衡量：也就是打造公司的科技基礎。

I don't know a way to create a horizontal P&L for a data ecosystem or the move to the cloud. So there are some things that we do that we work backwards from what is the bone structure that we need to win and we go out and build that.

我不知道要如何為一個數據生態系，或是上雲（move to the cloud），建立一張橫向損益表（P&L）。所以我們做的一些事情，是從「要贏所需的骨架結構（bone structure）」倒推，然後出去把它建起來。

And we are seeing that while it's going to be impossible to measure the return on some of these things. I think if there were a way to do it over time, it would turn out to be the most high-yielding investment that we've ever made. So it's a bit of a tough way to make a living for Capital One and for our investors. But I think it's a key reason. This approach, Moshe, is a key reason we're here today and a central reason that we have the opportunity set that we have, and I look forward to our investors enjoying the returns from patient commitment to this approach.

而我們也看到，雖然有些事情的報酬率將不可能被量化衡量。但我認為，如果有辦法隨時間去衡量，它最終會證明是我們做過報酬最高的投資。所以，對 Capital One 以及我們的投資人而言，這是一種有點艱難的經營方式。但我認為這是關鍵原因。Moshe，這種做法是我們今天能站在這裡的關鍵原因之一，也是我們擁有目前這組機會集合（opportunity set）的核心原因；我也期待投資人能享受到對這種做法耐心承諾所帶來的回報。

That concludes our Q&A session and our call for this evening. Thank you very much for joining us on this call. Thank you for your interest in Capital One. Have a great evening.

我們今晚的問答環節與電話會議到此結束。非常感謝各位參與本次電話會議。感謝各位對 Capital One 的關注。祝各位今晚愉快。

Thank you. This concludes today's conference call. Thank you for participating. You may now disconnect.

謝謝。今天的電話會議到此結束。感謝各位參與。您現在可以斷線。