Allstate Corp (ALL) 2026 Q1 法說會逐字稿

Thank you for standing by. Welcome to Allstate's first-quarter earnings investor call. (Operator Instructions) As a reminder, please be aware this call is being recorded.

感謝您耐心等候。歡迎參加 Allstate 第一季財報投資人電話會議。(接線員指示) 提醒您，本通話將被錄音。

Now, I'd like to introduce your host for today's call, Allister Gobin, Head of Investor Relations. Please go ahead, sir.

現在，我想介紹今天會議的主持人：投資人關係主管 Allister Gobin。先生，請開始。

Good morning, everyone. Welcome to Allstate's first-quarter 2026 earnings call. Yesterday, following close of the market, we issued our news release and investor supplement, and posted related materials on our website at allstateinvestors.com.

各位早安。歡迎參加 Allstate 2026 年第一季財報電話會議。昨日收盤後，我們發布了新聞稿與投資人補充資料，並已將相關資料上傳至我們的網站 allstateinvestors.com。

Today, our management team will discuss how Allstate is creating shareholder value. Then we will open up the lines for your questions. As noted on the first slide of the presentation, our discussion will include non-GAAP measures for which reconciliations are provided in the news release and investor supplement. We will also make forward-looking statements about Allstate's operations. Actual results may differ materially from those statements, so please refer to our 2025 Form 10-K and other public filings for more information on potential risks.

今天，我們的管理團隊將說明 Allstate 如何創造股東價值。接著我們將開放線路回答各位提問。如簡報第一張投影片所述，我們的討論將包含非 GAAP 指標，其調節表已載於新聞稿與投資人補充資料中。我們也將就 Allstate 的營運發表前瞻性陳述。實際結果可能與這些陳述有重大差異，更多潛在風險資訊請參閱我們的 2025 年 Form 10-K 與其他公開申報文件。

Let's start with three of our recent advertisements, and then Tom will begin.

我們先播放三支近期廣告，接著由 Tom 開始。

(video playing)

(播放影片)

(video playing)

(播放影片)

(video playing)

(播放影片)

Good morning. Thank you for investing time in Allstate. I hope you enjoyed the three ads, which show how we are using marketing to accelerate growth. The first was our Mayhem ad, which is known to all of you. The second reinforces a simple message, Check Allstate first. And then third, which debuted this week, is our newest campaign: If it's important to you, it's important to Allstate, which demonstrates our commitment and our care for customers and the breadth of our offering.

各位早安。感謝您投入時間關注 Allstate。希望您喜歡這三支廣告，它們展示了我們如何運用行銷來加速成長。第一支是各位都熟悉的 Mayhem 廣告。第二支強化一個簡單訊息：先查 Allstate。第三支是本週首播的最新活動：「如果對你重要，對 Allstate 也重要」，展現我們對客戶的承諾與關懷，以及我們產品服務的廣度。

These same themes apply to investors. You can avoid mayhem by investing in Allstate, which has a proven ability to generate consistent results. If you -- you should call Allstate first if you're investing in protection companies. We are affordable, particularly at this P/E ratio. If it's important to shareholders, it's important to Allstate. We're going to touch on these same themes this morning.

同樣的主題也適用於投資人。投資 Allstate 可避免「mayhem」，因為我們具備經驗證、能持續產生穩健成果的能力。如果您要投資保障型公司，您——您應該先找 Allstate。以目前的本益比來看，我們的估值相當具吸引力。如果對股東重要，對 Allstate 也重要。今天早上我們也會觸及這些相同主題。

So let's review first-quarter results starting on slide 2. Allstate had excellent operating results in the first quarter. As you know, our strategy has two components that are shown on the left: increased personal and property liability market share and expanded protection provided to customers. On the right are performance highlights for the first quarter.

接著我們從第 2 張投影片開始回顧第一季結果。Allstate 第一季營運表現非常出色。如各位所知，我們的策略包含兩個面向，顯示在左側：提高個人與財產責任險的市占率，以及擴大提供給客戶的保障。右側則是第一季的績效重點。

An important part of today's conversation is that Allstate competes using a broad set of tools, not just lowering price. This enables us to maintain attractive margins while accelerating growth. We also broaden protection offerings for customers, investment income increased, and shareholders received higher dividends and accelerated share repurchases.

今天討論的一個重點是：Allstate 的競爭方式仰賴一整套工具，而不只是降價。這讓我們能在加速成長的同時維持具吸引力的利潤率。我們也擴大客戶保障方案，投資收益增加，股東獲得更高股利並加速股票回購。

The financial results are shown on slide 3. Total revenues increased to $16.9 billion, up 3% for the first quarter of 2025. Investment income increased nearly 10% to $938 million. The property and liability recorded combined ratio is 82, and the underlying combined ratio is 80.3, a 2.8 point improvement from the prior year.

第 3 張投影片顯示財務結果。總營收增至 169 億美元，較 2025 年第一季成長 3%。投資收益增加近 10%，至 9.38 億美元。財產與責任險的已入帳綜合率為 82，基礎綜合率為 80.3，較去年改善 2.8 個百分點。

Total policies in force increased by 2.5%, and property and liability policies in force increased by 2.3%. Net income was $2.4 billion, and adjusted net income was $2.8 billion, or $10.65 per diluted share. Net income return on equity was 48.4% over the last 12 months.

有效保單總數增加 2.5%，其中財產與責任險有效保單增加 2.3%。淨利為 24 億美元，調整後淨利為 28 億美元，或每股稀釋後盈餘 10.65 美元。過去 12 個月的淨利股東權益報酬率為 48.4%。

Slide 4 provides a construct to answer the question: how will you generate attractive returns by growing if that includes more affordable pricing? The answer is that while prices are extremely important, transformative growth has created a broad set of competitive levers to enable us to grow, as you can see on the left. More affordable prices are supported by lower expenses and effective claims processes.

第 4 張投影片提供一個架構，用來回答這個問題：若成長包含更具可負擔性的定價，你們要如何在成長的同時創造具吸引力的報酬？答案是：雖然價格極其重要，但轉型式成長已打造出一套廣泛的競爭槓桿，使我們能如左側所示持續成長。更可負擔的價格由更低的費用與有效的理賠流程所支撐。

We also use sophisticated analytics, new products, expanded benefits, and bundled offerings to better serve customers, compelling marketing and broad distribution increase new business, which fuels growth. This flywheel results in market share increases. Some examples are shown on the right. Affordable prices and lower expenses are enhanced with sophisticated pricing plans and better customer experiences.

我們也運用先進分析、新產品、擴增的保障內容，以及組合式方案來更好地服務客戶；具吸引力的行銷與廣泛的通路帶動新業務，進而推動成長。這個飛輪效應帶來市占率提升。右側列出一些例子。可負擔的價格與較低費用，透過更精密的定價方案與更佳的客戶體驗而進一步強化。

New products and benefits create value for customers. The Allstate brand affordable, simple, and connected products for both auto and home insurance are now available in 45 and 36 states respectively. The Custom 360 Auto and Homeowners Insurance products for independent agents are now available in 40 states. We also routinely expand or improve benefits. For example, we recently added free Allstate Identity Protection, so customers think beyond price, and we execute this strategy of broadening protection.

新產品與新保障內容為客戶創造價值。Allstate 品牌的「可負擔、簡單、互聯」汽車與住宅保險產品，目前分別已在 45 州與 36 州提供。提供給獨立代理人的 Custom 360 汽車與住宅保險產品，目前已在 40 州提供。我們也會例行性擴充或改善保障內容。例如，我們近期新增免費的 Allstate 身分保護（Identity Protection），讓客戶不只看價格，並落實我們擴大保障的策略。

Allstate agents bundle auto and homeowners insurance at high rates, making it easier for customers and lowering acquisition costs per policy. Marketing acquisition economics have improved this year. We distribute to Allstate agents, independent agents, company call centers, and over the web, which provides the right level of service for customers at the best value. In the first quarter, all distribution channels had increases in new business, and the total was a record, which increased growth.

Allstate 代理人以高比例將汽車與住宅保險打包銷售，讓客戶更便利，並降低每張保單的取得成本。今年行銷獲客的經濟效益已有改善。我們透過 Allstate 代理人、獨立代理人、公司客服中心以及網路進行銷售，為客戶以最佳價值提供適切的服務水準。第一季所有通路的新業務均成長，總量創下紀錄，帶動成長提升。

Mario will now cover how this translates into market share growth while earning attractive returns. Jess will then review more specifics on the Property-Liability business, and John's going to cover Protection Services, investments, and capital.

接下來 Mario 將說明這如何在賺取具吸引力報酬的同時，轉化為市占率成長。之後 Jess 會更具體地回顧財產—責任險業務，而 John 將說明保障服務、投資與資本。

Thanks, Tom. Let's start with the market share growth on slide 5. Starting on the left, Allstate increased auto insurance market share in 29 states in 2025 that comprise 57% of countrywide premiums. Looking down below, in the 29 states where share increased, policies in force increased by 4.3% over the prior year and outpaced vehicle registration growth in those states. That means we increased our share of insurable vehicles in those states, which we view as a better indicator of sustainable share growth than the traditional premium-based market share metric. In the remainder of the country, policies in force decreased by 0.5% versus an increase in vehicle registration of 0.6%. The decline is heavily impacted by two large states where we have intentionally been reducing share because of profitability challenges.

謝謝你，Tom。我們先從第 5 張投影片的市占率成長談起。先看左側，Allstate 在 2025 年於 29 個州提高了汽車保險市占率，這些州合計占全國保費的 57%。往下看，在市占率提升的 29 個州中，有效保單數較去年增加 4.3%，且增速超過這些州的車輛登記成長。這表示我們在這些州提高了可承保車輛的占比；我們認為，這比傳統以保費為基礎的市占率指標，更能反映可持續的市占率成長。在其餘地區，有效保單數下降 0.5%，而車輛登記則增加 0.6%。這項下滑主要受到兩個大型州的影響；由於獲利性挑戰，我們在這兩州刻意降低市占率。

If you look at which companies this growth comes from by dividing the market into the top five market share leaders and the rest of the market, slightly more comes from the medium-sized and smaller carriers. The broad set of competitive tools that Tom referenced also drives growth in homeowners insurance. Homeowners insurance market share grew in 83% of the US market. This was in 41 states which had policy in force growth of 4.1% in 2025 over the prior year. We have a broad competitive advantage over the companies we compete with in the homeowners insurance market, as demonstrated by our ability to profitably gain share.

如果將市場區分為市占率前五大領導者與其餘市場，以此觀察這些成長來自哪些公司，則略多一些的成長來自中型與較小型的保險公司。Tom 所提到的廣泛競爭工具組合，也推動了住宅屋主保險的成長。住宅屋主保險的市占率在美國市場的 83% 範圍內有所提升。這發生在 41 個州，這些州在 2025 年的有效保單（policy in force）較前一年成長 4.1%。我們在住宅屋主保險市場相較於競爭對手具備廣泛的競爭優勢，這可由我們能夠在獲利的前提下提升市占率的能力加以證明。

Moving to slide 6, Allstate's business model enables us to consistently generate strong returns. On the chart, the blue bars represent the auto insurance underlying combined ratio, which averaged 95 and 94 over the last 5 and 10 years, consistent with our mid-nineties target. There was obviously an increase in the combined ratio in 2022 post-pandemic, which required significant price increases, as shown by the light blue line in the middle of this chart. Since then, we have returned to levels at or below our mid-nineties target with more modest price increases needed to generate and sustain attractive returns. In the first quarter of 2026, rate changes were implemented in 39 states, which included a mix of both rate increases and decreases. These changes had a net overall neutral implemented rate impact across the book.

接著看第 6 頁投影片，Allstate 的商業模式使我們能夠持續產生強勁的報酬。圖表中，藍色長條代表汽車保險的基礎綜合率（underlying combined ratio），過去 5 年與 10 年的平均值分別為 95 與 94，符合我們以 90 多中段為目標的水準。顯然在疫情後的 2022 年，綜合率有所上升，因而需要大幅調漲保費費率，如圖表中間的淺藍色線所示。自那之後，我們已回到等於或低於 90 多中段目標的水準，且僅需較溫和的費率調整即可產生並維持具吸引力的報酬。在 2026 年第一季，我們於 39 個州實施了費率變更，包含調升與調降的組合。這些變更對整體保單組合的已實施費率影響，淨額而言大致中性。

Improving affordability will increase policy in force growth and raise shareholder value as long as the combined ratio continues to perform at or better than target levels. Let me note that these are underlying combined ratios that were reported for these years. As Jess will cover in a few minutes, favorable subsequent reserve development shows that results for several of these years are actually better than what is shown on the chart.

只要綜合率持續達到或優於目標水準，提升可負擔性將會帶動有效保單成長並提高股東價值。我補充說明，這些是各年度所揭露的基礎綜合率。正如 Jess 幾分鐘後將說明的，有利的後續準備金發展顯示，其中數個年度的實際結果其實優於圖表所呈現的水準。

Moving to slide 7, you can see a similar story in the homeowners insurance business, which also generates strong returns. Homeowners insurance over the last 5 and 10 years had a reported combined ratio of 93.5 and 92 respectively, and has generated underwriting income of $3.9 billion and $7.9 billion in those same periods.

接著看第 7 頁投影片，你可以在住宅屋主保險業務中看到類似的情況，該業務同樣能產生強勁報酬。住宅屋主保險在過去 5 年與 10 年的揭露綜合率分別為 93.5 與 92，並在相同期間分別創造了 39 億美元與 79 億美元的核保利益（underwriting income）。

In the first quarter, the combined ratio was 83.5. Average premiums increased 5.7% compared to the prior year quarter, keeping pace with loss costs. As you saw this quarter, we also posted the disclosure related to the placement of our comprehensive nationwide reinsurance program, which enhances the risk and return profile in the homeowners business by reducing capital requirements associated with catastrophe loss tail risk and dampening earnings volatility. The homeowners insurance business remains a competitive advantage and growth opportunity for Allstate. Now let me turn it over to Jess.

第一季的綜合率為 83.5。平均保費較去年同期增加 5.7%，與損失成本的增幅一致。如同你在本季所見，我們也揭露了關於配置我們全國性綜合再保計畫的資訊；該計畫透過降低與巨災損失尾端風險相關的資本需求並抑制盈餘波動，強化住宅屋主保險業務的風險與報酬結構。住宅屋主保險業務仍是 Allstate 的競爭優勢與成長機會。接下來我把時間交給 Jess。

All right. Thanks, Mario. Let's look at the Property-Liability results in total on slide 8. Auto insurance policy growth of 2.6% and homeowners insurance policy growth of 2.5% drove an increase of 2.3% in total policies in force and written premiums. Earned premiums increased by 5.5%. The Property-Liability combined ratio was 82.0% as both auto and homeowners insurance profitability was better than our targeted levels. This result was due to strong underlying performance as well as lower catastrophes and favorable prior year reserve releases. Excluding the benefit of reserve changes and lower catastrophes, the auto insurance underlying combined ratio was 89.5%, which is 1.7 points better than prior year. Property-Liability underwriting income was $2.7 billion in the first quarter.

好的。謝謝你，Mario。我們來看第 8 頁投影片的整體財產—責任險（Property-Liability）結果。汽車保險有效保單成長 2.6% 與住宅屋主保險有效保單成長 2.5%，帶動有效保單總數與簽單保費（written premiums）增加 2.3%。已賺保費（earned premiums）增加 5.5%。財產—責任險綜合率為 82.0%，汽車與住宅屋主保險的獲利能力皆優於我們的目標水準。此結果來自強勁的基礎表現，以及較低的巨災損失與有利的前期年度準備金釋放。若排除準備金變動與較低巨災損失的效益，汽車保險基礎綜合率為 89.5%，較去年改善 1.7 個點。第一季財產—責任險核保利益為 27 億美元。

Now, turning to slide 9, as Mario referenced in his comments, auto insurance profitability improved faster than original estimates in 2023 and 2024. The top of the stacked bar is the underlying combined ratio as originally reported. The green bars represent the impact of subsequent prior year reserve re-estimates. The light blue bars represent the adjusted underlying combined ratio, including the subsequent changes in our estimates of loss costs.

接著看第 9 頁投影片，如 Mario 在評論中提到的，汽車保險獲利能力在 2023 與 2024 年改善速度快於原先估計。堆疊長條的最上方是最初揭露的基礎綜合率。綠色長條代表後續對前期年度準備金重新估計的影響。淺藍色長條代表調整後的基礎綜合率，已納入我們對損失成本估計的後續變動。

As you can see, prior year losses developed more favorably than originally estimated. Reserving is an iterative process with strong governance and oversight. We use consistent practices, multiple analytical methods, and include external reviews by independent actuaries to ensure reserve adequacy.

如你所見，前期年度損失的發展情況比原先估計更為有利。準備金提列是一個在強健治理與監督下進行的反覆迭代流程。我們採用一致的作法、多種分析方法，並納入獨立精算師的外部審查，以確保準備金充足。

As more claims settle, however, estimates each year are revised to reflect actual loss experience. In recent quarters, actual loss experience has outperformed initial expectations. This results in the release of reserves from prior years. The auto combined ratio in 2023 is now estimated at 95.4%, and 2024 is estimated at 90.0%. Auto insurance profitability improved faster than originally estimated.

然而，隨著更多理賠案件結案，每年的估計都會修正，以反映實際損失經驗。在最近幾個季度，實際損失經驗優於初始預期。因此會釋放前期年度的準備金。目前估計 2023 年汽車保險綜合率為 95.4%，2024 年估計為 90.0%。汽車保險獲利能力改善速度快於原先估計。

Slide 10 highlights how we expect to continually improve our strong performance and enhance competitive position. Transformative growth built a comprehensive competitive model. This included new software and adapted legacy systems to build a connected technology ecosystem. The system enables the use of artificial intelligence to improve customer experience and lower costs. We're leveraging this technology platform in building Allstate's large language intelligent ecosystem, which we call ALLIE, to harness the power of agentic AI. With that, I'll turn it over to John.

第 10 頁投影片重點說明我們預期如何持續提升強勁表現並強化競爭地位。轉型式成長建立了一個全面性的競爭模型。其中包含導入新軟體並調整既有系統，以打造互聯的科技生態系。該系統使我們能運用人工智慧改善客戶體驗並降低成本。我們正運用此技術平台建構 Allstate 的大型語言智慧生態系，我們稱之為 ALLIE，以發揮代理式 AI（agentic AI）的力量。接下來我把時間交給 John。

Thanks, Jess. Good morning, everyone. Moving to slide 11, the protection services business continue to grow profitably. This segment is comprised of five businesses shown on the left: protection plans, dealer services, roadside, Arity, and identity protection. The largest business in this segment is Allstate Protection Plans, which grew revenue 13.5% versus the prior year quarter. This business provides protection for mobile phones, consumer electronics, major appliances, and furniture.

謝謝，Jess。各位早安。接著看第 11 頁投影片，保障服務（protection services）業務持續以獲利方式成長。此分部由左側所示的五項業務組成：保障方案（protection plans）、經銷商服務（dealer services）、道路救援（roadside）、Arity，以及身分保護（identity protection）。此分部中規模最大的業務是 Allstate Protection Plans，其營收較去年同期成長 13.5%。該業務為手機、消費性電子產品、大型家電與家具提供保障。

Protection plans generated $41 million in adjusted net income for the first quarter, down slightly due to higher claims costs. Arity is our mobile intelligence business. The higher loss this quarter reflects restructuring charge related to a reduced employee count. In total, protection services businesses increased revenue 7.2% from the first quarter of 2025 and generated $47 million in adjusted net income.

第一季保障方案創造 4,100 萬美元的調整後淨利（adjusted net income），因理賠成本較高而略為下降。Arity 是我們的行動智慧（mobile intelligence）業務。本季較高的虧損反映了因員工人數縮減而產生的重整費用。整體而言，保障服務業務的營收較 2025 年第一季增加 7.2%，並創造 4,700 萬美元的調整後淨利。

Let's turn to slide 12 to discuss investment portfolio. Investment income of $938 million increased $84 million, or 9.8% compared to the prior year quarter. As shown on the chart on the left, net investment income has grown as the portfolio grew. Since the first quarter of 2024, portfolio book value has increased 24% or approximately $17 billion. The increase reflects higher average investment balances from a 15% increase in earned premiums, strong underwriting income, and improved fixed income yields. The table on the right side highlights the strength and consistency of returns across asset classes.

接著看第 12 頁投影片，討論投資組合。投資收益為 9.38 億美元，較去年同期增加 8,400 萬美元，或 9.8%。如左側圖表所示，隨著投資組合規模成長，淨投資收益也隨之增加。自 2024 年第一季以來，投資組合帳面價值增加 24%，約 170 億美元。此增幅反映了因已賺保費增加 15% 而帶來的較高平均投資餘額、強勁的核保利益，以及固定收益殖利率的改善。右側表格突顯各資產類別報酬的強度與一致性。

Over the last 12 months, the portfolio generated a 4.2% return. Fixed income results over the last five years are top quartile. Returns in our performance-based portfolio have been below longer-term historic averages over the last 1 and 3 years at 7.6% and 5.9% respectively, but remain above industry benchmarks. These results underscore the effectiveness of our active investment management approach. As a result, we increased the capital allocated to the investment portfolio in the first quarter, some of which is carried at the holding company.

在過去12個月中，投資組合產生了4.2%的報酬率。過去五年的固定收益表現位居同業前四分位。我們以績效為基礎的投資組合在過去1年與3年的報酬率分別為7.6%與5.9%，低於較長期的歷史平均水準，但仍高於產業基準。這些結果凸顯了我們主動投資管理方法的有效性。因此，我們在第一季提高了配置至投資組合的資本，其中一部分由控股公司持有。

Let's move to slide 13 to show that proactive capital management creates shareholder value. Allstate deploys capital in multiple ways, which are shown on the left axis. Organic growth, enhancing existing businesses, growth acquisitions, and cash provided to shareholders. Using capital for organic growth leverages Allstate's capabilities and market presence with well understood and attractive risk and return opportunities. This is why we're focusing on increasing market share in the Property-Liability business.

接著我們看第13頁投影片，說明主動的資本管理如何創造股東價值。Allstate 以多種方式運用資本，如左側軸所示。包括有機成長、強化既有業務、成長型併購，以及回饋股東的現金。將資本用於有機成長，可運用 Allstate 的能力與市場地位，掌握風險與報酬特性清楚且具吸引力的機會。這也是我們聚焦於提高財產—責任險業務市占率的原因。

In addition, increasing market share should raise valuation multiples. Over the last 3 years, $3 billion of economic capital was utilized to support premium growth. As we just discussed, Allstate also deploys capital to support the investment portfolio to generate attractive risk-adjusted returns. Capital is also used to strengthen existing business, the businesses, such as investments we made in our technology ecosystem or enhancing our independent agent business through the acquisition of National General.

此外，提高市占率應可提升估值倍數。過去3年中，動用了30億美元的經濟資本以支持保費成長。如同我們剛才討論的，Allstate 也會配置資本以支持投資組合，創造具吸引力的風險調整後報酬。資本也用於強化既有業務，例如我們對科技生態系的投資，或透過收購 National General 來強化我們的獨立代理人業務。

SquareTrade was a growth acquisition that leveraged the Allstate brand and capabilities. It also expanded protection offerings to execute the second part of our strategy and brought strong retail distribution partnerships. Since it was acquired, revenues have increased eightfold and the business generated $175 million of adjusted net income over the last 12 months. Allstate also has a long track record of returning capital to shareholders. In the first quarter, $881 million was returned to shareholders through repurchases and dividends. We completed the former $1.5 billion share repurchase program and launched a new $4 billion share repurchase program, accelerating the pace of repurchases. $3.6 billion remains on the current share repurchase authorization, which represents approximately 40% of holding company assets as of March 31 and 7% of outstanding shares.

SquareTrade 是一項成長型併購，運用了 Allstate 的品牌與能力。它也擴大了保障型產品供給，以執行我們策略的第二部分，並帶來強大的零售通路合作夥伴關係。自收購以來，營收成長了八倍，且該業務在過去12個月創造了1.75億美元的調整後淨利。Allstate 也長期以來持續將資本回饋給股東。第一季透過庫藏股回購與股利，回饋股東8.81億美元。我們完成了先前15億美元的股票回購計畫，並啟動新的40億美元股票回購計畫，加快回購步伐。截至3月31日，目前回購授權尚餘36億美元，約相當於控股公司資產的40%，以及流通在外股數的7%。

As an interesting observation, if you bought all of Allstate 10 years ago, you would have received 99% of the purchase price back in cash and would have a company that generated $12 billion in net income over the last 12 months. Wrapping up on slide 14, in summary, Allstate's broad set of competitive levers delivered strong results in the first quarter.

作為一個有趣的觀察，如果你在10年前買下整個 Allstate，你將以現金形式收回相當於購買價格99%的金額，並且仍持有一家在過去12個月創造了120億美元淨利的公司。在第14頁投影片做個收尾：總結而言，Allstate 廣泛的競爭槓桿在第一季帶來了強勁的成果。

Now, let's move to questions.

現在，我們進入提問環節。

Michael Zaremski, BMO.

Michael Zaremski，BMO。

Hey, good morning. This is Jack on for Mike. Maybe just first on the pricing outlook, given how strong reported loss ratios are across your portfolio, I'm wondering how you're thinking about the opportunity to lean in more aggressively on pricing this year, and does that calculus differ materially across auto, homeowners and bundled customers?

嗨，早安。我是 Jack，代替 Mike 發言。先就定價展望提問：鑑於你們整體投資組合所呈報的損失率表現相當強勁，我想了解你們如何看待今年在定價上更積極加碼的機會？而這樣的考量在汽車險、住宅險以及綁單客戶之間，是否有顯著差異？

I would go back to the slide we talked about in terms of growing. We have a wide range of ways in which we grow. Price is certainly important, but it's not the only one. I know this is a question for many of you, let's spend a minute to. Because it's what you describe. We do it obviously by product, we do it by state, we do it by coverage. It's highly complicated. If we think bundled customers, lower acquisition costs, we give them a discount if they bundle. Yes, we do all that. Let me go up. Price is obviously important, and it's a key driver of profitability. As a result, we built this system called operational levers, organizational accountability, and sophisticated analytics.

我會回到我們談到成長的那張投影片。我們有多種成長方式。價格當然重要，但不是唯一因素。我知道這是你們許多人關心的問題，我們花一點時間說明。因為正如你所描述的。我們顯然會按產品、按州別、按保障範圍來做。這非常複雜。若以綁單客戶來看，獲客成本較低，我們也會在綁單時給予折扣。是的，我們都會做。我再往上說。價格顯然重要，也是獲利能力的關鍵驅動因素。因此，我們建立了一套稱為「營運槓桿、組織問責，以及精密分析」的系統。

Our goal of course is to earn attractive margins and grow. There's always a plan on prices that looks forward 6 to 12 months. We're not going to talk about what that plan is here because it's competitive and it changes all the time. It's based on what operational levers we think we can pull. Jess will describe the system for you and give you a couple examples of how it works. The conclusion, however, is that the system works. It works for auto and it works for homeowners, and you can see that on slide 6 and 7. Our auto combined ratio is 94%-95% over the last 5 and 10 years. Homeowners combined ratio 92%-93.5% over the last 5 and 10 years. The system itself works.

我們的目標當然是取得具吸引力的利潤率並成長。在價格方面，永遠都有一個向前看6到12個月的計畫。我們不會在這裡談那個計畫的內容，因為這涉及競爭，而且也一直在變。它是基於我們認為可以運用哪些營運槓桿。Jess 會為各位說明這套系統，並舉幾個運作方式的例子。不過結論是：這套系統有效。它對汽車險有效，對住宅險也有效，你們可以在第6與第7頁投影片看到。我們的汽車險綜合成本率在過去5年與10年為94%至95%。住宅險綜合成本率在過去5年與10年為92%至93.5%。這套系統本身是有效的。

While price is important, it's just one component. Jess, why don't you talk about how it works here, and then give a couple examples.

雖然價格很重要，但它只是其中一個組成部分。Jess，你不如在這裡談談它如何運作，並舉幾個例子。

Got it. We think about the system like a cube that has three elements, and Tom alluded to the three elements. You have operational levers, you have advanced analytics, and then organizational roles and responsibilities. It's a bit like a Rubik's Cube, where it gives us multiple ways to both identify and address profit and growth opportunities that we have. What I'll do quickly is go through each component and I'll give a couple examples of what's going on, a couple state examples of how the system works. If you start with the operational lever element of our cube, we kind of covered this on slide four. Tom went through it. You have new products, broad distribution, marketing. Effectively, we employ these operational levers at the state individual market and product level. It's very granular.

了解。我們把這套系統想像成一個由三個要素構成的立方體，而 Tom 也提到了這三個要素。包括營運槓桿、進階分析，以及組織角色與責任。它有點像魔術方塊，讓我們能以多種方式同時辨識並處理我們所擁有的獲利與成長機會。我會快速走過每個組成部分，並舉幾個目前正在進行的例子，以及幾個州別的例子，說明這套系統如何運作。如果從立方體中的「營運槓桿」要素開始，我們在第4頁投影片已大致涵蓋。Tom 也逐一說明過。你有新產品、廣泛的通路、行銷。實際上，我們會在州別、個人市場與產品層級運用這些營運槓桿。顆粒度非常細。

If I move to the advanced analytics element, you have a highly sophisticated rating plans that have billions of price points per state. We analyze data by sub-market within each state and by product, by coverage, by risk segment. We link that between the signals that we're seeing in current claims trends to price at a very granular level. We're bringing again this interconnected system together. We have marketing analytics that are terrific. They enable us to price lead purchases in real time, determine effectiveness of programs by media channel and message. The claims team is using a massive amount of data to assess the effectiveness of controlling severity and executing the claims function. We have a centralized reserving team of course, and we've talked about that. That's separate from our actuarial pricing team.

接著談「進階分析」要素：我們有高度精密的費率方案，每個州都有數十億個價格點。我們會在各州內按次市場、按產品、按保障範圍、按風險區隔來分析資料。我們把目前理賠趨勢中看到的訊號，與非常細緻層級的定價連結起來。我們再次把這個相互連結的系統整合在一起。我們的行銷分析非常出色。它們讓我們能即時為名單購買定價，並依媒體渠道與訊息判斷方案成效。理賠團隊運用大量資料來評估控制損失嚴重度的成效，以及理賠職能的執行情況。當然，我們也有一個集中的準備金團隊，我們也談過。該團隊與我們的精算定價團隊是分開的。

That gives us another set of eyes on loss costs and loss cost trends. The point of all this is that we have a lot of people looking at profitability and growth from a number of different perspectives through the advanced analytic lens. The final element, as Tom mentioned, was organizational roles and accountability. We have a matrix organization structure that enables us to bring all of our expertise to bear to decide how to pull various levers in this system. That includes price changes in total or by territory or by coverage or customer risk segment. It includes adjusting underwriting guidelines. Another dimension to that would be marketing investment. We can look at price, number of sales leads to purchase by market, and then determine distribution priorities alongside those other decisions.

這讓我們能以另一組視角來檢視損失成本與損失成本趨勢。這一切的重點在於，我們有許多人透過先進分析的視角，從多種不同觀點來檢視獲利能力與成長。最後一個要素，如 Tom 所提到的，是組織角色與問責。我們採用矩陣式組織架構，使我們能整合所有專業能力，來決定在這個系統中如何拉動各種槓桿。這包括整體或按地區、按保障範圍、或按客戶風險分群的價格調整。也包括調整核保指引。另一個面向則是行銷投資。我們可以按市場檢視價格、從銷售線索到購買的轉換數量，並在其他決策的同時，決定通路優先順序。

It's a system that's working together, again, like a Rubik's Cube to drive profitable growth. The overall team, as we look at it, includes state managers that are responsible for profitability by product line, territory, and coverage. We have a chief actuary who oversees analytics, pricing trends across the country and by state, and has a research and development function. We have go-to-market teams that are out there each day bringing all their expertise and all this expertise together to manage growth and profitability by local market. Then we have distribution leads for Allstate agents, independent agents in our direct operations who can assess and evaluate performance on a real-time basis. They can expand or shrink distribution, and set priorities and compensation to make sure, again, that we're optimizing across the system.

這是一個協同運作的系統，再次就像魔術方塊一樣，用來推動具獲利性的成長。從我們的角度來看，整體團隊包括州級經理，負責按產品線、地區與保障範圍的獲利能力。我們有一位總精算師，負責監督分析、全國及各州的定價趨勢，並具備研發職能。我們有市場推進（go-to-market）團隊，每天在外整合他們的專業與所有這些專業能力，按在地市場管理成長與獲利。接著我們有通路主管，涵蓋 Allstate 代理人、以及我們直營業務中的獨立代理人，他們能即時評估與衡量績效。他們可以擴大或縮減通路，並設定優先事項與酬金，以確保我們再次能在整個系統中達到最佳化。

The three elements work together in a continuous planning cycle, is the way that I think of it. We create a forward-looking plan that looks at expected rate changes for the next 6-12 months by state, by line, by company, as Tom referenced. It factors in things like likely regulatory timing and what the response will be, and we build up a countrywide matrix then of underlying profitability and growth so that we can evaluate the forward-looking trajectory. It aligns the execution of all of the operational levers with the goal of earning attractive returns and growing in 2027 and 2028. To make turning the dimensions of this Rubik's Cube look like come alive a bit, I thought I would talk about a couple of examples.

我認為這三個要素是在一個持續的規劃循環中共同運作。我們制定一個前瞻性計畫，如 Tom 所提到的，按州、按險種、按公司，檢視未來 6–12 個月預期的費率變動。它會納入例如可能的監管時程以及可能的回應，我們因此建立一個全國性的矩陣，呈現基礎獲利能力與成長，以便評估前瞻性的走勢。它使所有營運槓桿的執行與目標對齊，也就是在 2027 與 2028 年賺取具吸引力的報酬並實現成長。為了讓轉動這個魔術方塊的各個面向更具體一些，我想談幾個例子。

In states where we have share that we would say is below our national average and our underlying combined ratio is better than target, say we're running an 88 underlying combined ratio, state managers will identify an opportunity to lower rates with an eye towards staying within those targeted ranges, in coming quarters and in coming years. It's a forward-looking view so that we change rates in a sustainable way. They then work within the system that I referenced to utilize the broad set of tools that we discussed in our prepared remarks to drive profitable growth by market. That's optimizing distribution. It's working with the marketing team to make sure that where we have opportunity to grow, we're leaning into that.

在一些州，我們的市占率低於全國平均，而我們的基礎綜合率優於目標，例如基礎綜合率為 88，州級經理會辨識到降費率的機會，同時著眼於在未來幾季與未來幾年仍維持在目標區間內。這是一種前瞻性的觀點，讓我們以可持續的方式調整費率。接著他們會在我提到的系統內運作，運用我們在事先準備的發言中討論的廣泛工具組合，按市場推動具獲利性的成長。這包括最佳化通路。也包括與行銷團隊合作，確保在我們有成長機會的地方，我們會加大投入。

On the other hand, the other state example would be a state where our underlying combined ratio is above target or on a trajectory to go above our target. We begin taking modest rate increases to get ahead of the trend. If needed, we'll restrict new business through underwriting guidelines and other operational levers, again, that we have to make sure that we manage profitability in that state. In states where we don't have ASC, now we do have ASC in 40-plus states at this point, we'll limit new business until that product is available because we want the most contemporary and most accurately priced product in market. We'll wait, make sure that ASC gets approved and then relook at growth on a forward-looking basis.

另一方面，另一個州的例子會是：我們的基礎綜合率高於目標，或其走勢將高於目標。我們會開始採取溫和的費率上調，以提前因應趨勢。如有需要，我們會透過核保指引與其他營運槓桿來限制新業務，以確保我們能管理該州的獲利能力。在尚未導入 ASC 的州——目前我們已在 40 多個州導入 ASC——我們會限制新業務，直到該產品可供使用，因為我們希望市場上提供的是最新、且定價最精準的產品。我們會等待、確保 ASC 獲得核准，然後再以前瞻性的方式重新檢視成長。

We get the best product in market and, again, look across the system to make sure that, we're appropriately adjusting for a state that is, not meeting our targeted returns. To make the couple examples come alive, I thought I would just end with the system at work. You saw in the supplement that we changed auto rates in 39 locations, and that netted to effectively no change, in rate. If you peel that back, there were 23 states where we lowered rates. There were 16 states where we increased rates. Because of our rating sophistication and segmentation, 10 of those states, we did both. We had an increase and a decrease. This is more than just a high-level analysis.

我們要在市場上提供最好的產品，並再次從整個系統的角度檢視，以確保我們對於未達成目標報酬的州，做出適當調整。為了讓這幾個例子更具體，我想最後用系統實際運作的情況作結。你在補充資料中看到，我們在 39 個地點調整了汽車保險費率，而淨效果是費率基本上沒有變動。如果進一步拆解，其中有 23 個州我們調降了費率。有 16 個州我們調升了費率。由於我們在定價上的精細化與分群能力，在其中 10 個州，我們兩者都做了。我們同時有上調與下調。這不只是高層次的分析。

It shows the depth and the breadth of what we're doing to pull the operational levers and all the levers that are in the Rubik's Cube to optimize and deliver profitable growth.

這顯示了我們為了拉動營運槓桿、以及魔術方塊中的所有槓桿，以達到最佳化並交付具獲利性的成長，所做工作的深度與廣度。

That's helpful perspective. Thank you. Maybe just a follow-up on California, where they recently came out with some reforms to the intervenor process. I guess wondering, does that change along with other recent reforms, there is potentially a game changer longer term, especially on the homeowner side, where I think historically you've been reluctant to grow market share?

這個觀點很有幫助。謝謝。也許再追問一下加州：他們最近針對介入人（intervenor）程序推出了一些改革。我想了解的是，這項變動加上其他近期改革，長期而言是否可能成為改變遊戲規則的因素，特別是在住宅險端；我認為過去你們在那邊對於擴大市占率一直較為保守？

We believe that California still has a significant number of changes to make before the homeowners market will be accurately priced with decent availability for consumers.

我們認為，加州在住宅險市場能以合理供給、並以準確價格反映之前，仍需要進行相當多的變革。

Thank you.

謝謝。

Joshua Shanker, Bank of America.

Joshua Shanker，美國銀行。

Yeah. Thank you, for taking my question. In the first quarter, you had about $840 million of net favorable prior year development in the auto line. Obviously, I would imagine the majority of that comes from last year, which tells me you made a lot more money in auto last year than the combined ratio indicates. It also arguably suggests that, year-over-year, the margins are deteriorating. I mean, they will. They're incredible right now. They have to deteriorate at some point. I'm wondering if you can talk about the trajectory of what you think is happening right now to help us better understand that.

是的。謝謝你回答我的問題。第一季，你們在汽車險線別有大約 8.4 億美元的前期年度（prior year）準備金發展淨有利（net favorable development）。顯然，我想其中大部分來自去年，這表示你們去年在汽車險賺到的錢比綜合率所呈現的還要多。這也或許意味著，年對年而言，利潤率正在惡化。我的意思是，它們會惡化。現在的利潤率非常驚人。它們在某個時間點必然會惡化。我想請你談談你認為目前正在發生的走勢，幫助我們更好地理解。

Josh, if you go to slide nine. Slide nine. You can see how we spread that. Actually, most of the change as it relates to the combined ratio came in 2023 and 2024. Very little in 2025. That's in part because 2025 hasn't completely developed. Like, we make these changes, we obviously do an estimate. We start settling claims. As we settle claims, we figure out what we're having to pay people, figure out how severe they are, and then we adjust our estimates. We obviously overshot the mark in 2023 and 2024. We have not concluded that for 2025. We think our reserves are properly stated. I'd also point out we really didn't overshoot the mark much in 2023. It happened to be those really concentrated in those two years.

Josh，如果你看第九頁投影片。第九頁。你可以看到我們如何分攤（spread）那個影響。其實，與綜合率相關的大部分變動發生在 2023 年與 2024 年。2025 年的變動很少。部分原因是 2025 年尚未完全發展（develop）。例如，我們做這些變更時，顯然是先做估計。我們開始理賠結案。隨著理賠結案，我們會了解實際需要支付給保戶的金額、了解損失嚴重度，然後調整我們的估計。我們在 2023 年與 2024 年顯然高估了。我們尚未對 2025 年做出這樣的結論。我們認為我們的準備金提列是適當的。我也要指出，我們在 2023 年其實並沒有高估太多。只是剛好高度集中在那兩年。

Going forward, we feel good about profitability. We've been able to earn better than industry average combined ratios in auto insurance for a long time, and we expect to continue to do that. Will we still be at 89? I think when you look at the math on it, the extent we can drive growth and give up some margin that works to improve the shareholders' valuation multiples. That said, like we're okay earning what we have right now. Like we think we're competitive in the market, but we think we can grow faster.

展望未來，我們對獲利能力感到很有信心。我們在汽車保險長期以來都能取得優於業界平均的綜合率，我們預期也會持續如此。我們還會維持在 89 嗎？我認為從數學上來看，我們能在多大程度上推動成長、並在一定程度上讓出一些利潤率，這反而有助於提升股東的估值倍數。話雖如此，我們也能接受目前所賺到的水準。我們認為自己在市場上具競爭力，但也認為我們可以成長得更快。

Obviously, 2023 was a very strange year. Is there something in your process that says that you want to be more conservative on the most recent accident years, that the confidence interval on your reserving is more conservative for the most recent year and that programmatically, if you're doing things correctly, you would have this type of reserve release action going forward into 2027 as we look back to 2025?

顯然，2023 年是非常奇怪的一年。在你們的流程中，是否有什麼機制會讓你們對最近的出險年度採取更保守的做法，也就是說，你們在最近年度的準備金估計信賴區間會更保守；而且如果你們以程式化方式正確執行，當我們回頭看 2025 年時，是否會在往後到 2027 年都出現這種準備金釋放的動作？

No, we apply the same statistical standards to every year. I would say one of the things we're hopeful about is with advanced computing power that we can increasingly get more specific on what's in the reserves. Of course, the reserves are like you have a bunch of losses in a year, then you have to. You say, well, how much do we pay out? Then you're kind of doing it on the residual value basis. What we paid out determines what we have left for all the claims that we still have yet to settle. We think with advanced analytics, we may be able to get another angle on just looking at all the individual cases, which is really complicated. You get 900-page medical files.

沒有，我們對每一年都採用相同的統計標準。我會說我們抱持希望的一點是，隨著先進運算能力的提升，我們可以愈來愈具體地辨識準備金裡面到底包含什麼。當然，準備金就像是：某一年發生了一堆損失，接著你就必須……你會說，好，那我們要賠付多少？然後你某種程度上是以剩餘價值的基礎在做估計。我們已經賠付出去的金額，決定了我們為所有尚未結案的理賠案件還剩下多少準備金。我們認為透過先進分析，我們或許能從另一個角度切入，不只是看整體，而是檢視所有個別案件——這真的非常複雜。你會拿到 900 頁的醫療檔案。

You get like lots of stuff to try and figure out what that claim will settle at. It's the same process, same standards, and I would say always getting better, as we go forward. Of course, what you never really know is what's going to happen, with legal trends or anything else.

你會拿到很多資料，試著判斷那件理賠最後會以多少金額和解。流程一樣、標準一樣，而且我會說我們會在前進的過程中持續變得更好。當然，你永遠無法真正知道的是，法律趨勢或其他因素接下來會發生什麼。

Thank you for indulging all my questions.

謝謝你耐心回答我所有的問題。

Alex Scott, Barclays.

Alex Scott，巴克萊。

Hi. Thanks for taking the question. First one I wanted to ask you about, just prioritization of the holdco cash, which has grown to a pretty significant amount at this point. How would you think about prioritizing that? Are there different verticals within services that you'd look to expand or other things beyond obviously the larger buyback that you've been doing?

嗨。謝謝讓我提問。第一個我想問的是，關於控股公司（holdco）現金的優先順序配置；目前這筆現金已經累積到相當可觀的規模。你們會如何思考其優先順序？在服務業務內是否有不同的垂直領域會想擴張，或是除了你們一直在做、而且規模更大的庫藏股回購之外，還有其他用途？

Well, it's an important question, Alex. Let me try to build up a little, start a little bit above where John went and then talk about some specifics underneath that. John, feel free to jump in here. The first thing on capital is you got to get a great return on what you got. We had a 44% adjusted net income return on capital. All of our capital, there's no hiving stuff off, no separate closed books or anything like that. We got a 44%, so that's a good thing. When you look at the S&P 500, it's probably half of that. I don't know what it is this quarter, but typically it's in the low 20s.

Alex，這是個重要的問題。我先把脈絡往上拉一點，從比 John 剛才提到的更高層開始鋪陳，再往下談一些具體細節。John，你也可以隨時補充。資本配置的第一件事，是你必須在既有資本上取得很好的報酬。我們的調整後淨利對資本報酬率是 44%。我們所有的資本都是整體運用，沒有把任何東西切出去，沒有獨立的封閉帳簿或類似安排。我們拿到 44%，這是好事。如果你看標普 500，可能只有這個的一半。我不知道這一季是多少，但通常是在 20% 出頭。

We feel good about that return, particularly when you're buying it at this kind of P/E. You say, okay, well, what else can we do? John went through the order. Organic growth, you're just leveraging your existing capabilities, great scale to it. You just pump more volume through the system. Obviously, that's something we're focused on, but you got to make money at it. You don't want to end up losing money or give yourself a short-term sugar high of growth and a long-term hangover called low profitability. We manage that, as Jess talked about, very aggressively in the Property-Liability business.

我們對這樣的報酬率感到很滿意，尤其是在以這樣的本益比（P/E）買進的情況下。你會說，好，那我們還能做什麼？John 已經把順序講過了。有機成長，就是運用你既有的能力，且具備很好的規模效益。你只是把更多的量灌進系統。顯然這是我們聚焦的方向，但你必須在這上面賺錢。你不希望最後變成賠錢，或是為了短期成長的「糖分刺激」，換來長期的「宿醉」——也就是低獲利能力。正如 Jess 談到的，我們在產險（Property-Liability）業務上非常積極地管理這件事。

We also think there's plenty of ways we can expand and leverage our existing capabilities, whether that's John talked about, expanding our property liability businesses, or our investment using our investment capabilities, which we put a little more money into earlier this year because we think we're good at it, and the results show we're good at it, and we thought we saw some opportunities in the marketplace. From an enterprise risk and return perspective, we had room to do that. We look at it in total, we manage capital. Then there's a variety of other ways we can do it. In general, we look at it and say, we have to be a better owner of a business. Like, why would our ownership make this business better? That's where you look at your growth stuff.

我們也認為有很多方式可以擴張並槓桿化我們既有的能力：不論是 John 提到的，擴大我們的產險業務；或是運用我們的投資能力來做投資——我們今年稍早在投資上多投入了一些資金，因為我們認為我們很擅長，結果也證明我們很擅長，而且我們覺得在市場上看到了一些機會。從企業整體風險與報酬的角度來看，我們有空間這麼做。我們是以整體視角來看，並管理資本。接著還有各式各樣其他方式可以運用資本。總體而言，我們會看著它並說：我們必須成為一家企業更好的擁有者。也就是說，為什麼由我們來持有，能讓這門生意變得更好？這就是你去看成長相關事項的地方。

When you look at when we bought SquareTrade, putting our brand on it and that kind of retail distribution, we really ran the table on that business. Feel really good about it. Those don't come along that often, but you're always looking for ways in which you can enhance your capabilities. John, anything you would add to that?

回頭看我們收購 SquareTrade 的時候，把我們的品牌放上去，再加上那種零售通路的分銷，我們在那門業務上幾乎是全面勝出。我們對此感到非常滿意。這種機會不常出現，但你會一直尋找能夠強化自身能力的方法。John，你還有什麼要補充的嗎？

I think you covered most of it, Tom. Yeah, maybe just a couple things to point out that it really is a system decision. We're looking both outside of the firm at opportunities, but then also in the firm, what's the best trade-off of how that mix comes together. I would point out that sometimes it's harder to see some of the investments that we're making, such as in technology, or even in an investment portfolio. I think those can be fairly consumptive in terms of capital. It might be more difficult for you actually to see that versus a transaction.

Tom，我想你已經涵蓋大部分了。是的，也許再指出幾點：這確實是一個系統性的決策。我們一方面在公司外部尋找機會，但同時也在公司內部評估：在各種選項的組合之下，什麼樣的取捨最合適。我也想指出，有時候你們比較不容易看見我們正在做的一些投資，例如科技方面，甚至是投資組合上的投入。我認為這些在資本占用上可能相當可觀。相較於一筆交易案，你可能更難實際看出這些投入。

I guess I'd end up on the fact that I know some of you picked up on it, and it's in the queue. But we actually accelerated our share repurchase program throughout the quarter. And that wasn't just a one-time thing. We continued to accelerate it. One way to look at repurchases is what the quantum is, but also the pace matters too.

我最後想落在一點：我知道你們當中有些人已經注意到了，而且這也在提問序列裡。但我們其實在本季加速了我們的股票回購計畫。而且那不只是一次性的。我們持續在加速。看回購有一種方式是看回購的量有多少，但回購的速度同樣重要。

Got it. All very helpful. Second question, I actually want to circle back on artificial intelligence specifically. I know you guys have had a strategy over time to improve the expense ratio so you could get even more competitive in the market and spur some growth. Could you talk about how AI expands on that, what you're planning to do, and sort of how you see yourself positioned relative to some of your peers one of which I think has begun to roll it out more aggressively and reduce their workforce more.

了解。都非常有幫助。第二個問題，我其實想回到人工智慧（AI）這個主題。我知道你們長期以來有一個策略，是改善費用率，讓你們在市場上更具競爭力並帶動一些成長。你能談談 AI 如何在這方面擴展、你們打算怎麼做，以及你們相對於一些同業（其中一家我認為已開始更積極地推行並進一步縮減人力）你們覺得自己的定位如何嗎？

Let me start with the competitive position, and then come back up to how we're doing it, and I'll focus on both, expenses, AKA generative AI, and effectiveness called agentic AI. I think it's really hard to tell where everybody is. Everybody's out doing something. We don't talk about everything we're doing because we don't want everybody to know what we're doing. We'll let them see in the marketplace. I think it's hard. What I can say is that, from our standpoint, our capabilities continue to grow exponentially.

我先從競爭地位談起，接著再回到我們是如何做到的；我會同時聚焦在兩個面向：費用（也就是生成式 AI）以及被稱為代理式 AI 的效能。我認為很難判斷每個人目前進展到哪裡。大家都在做一些事情。我們不會談我們正在做的所有事情，因為我們不希望每個人都知道我們在做什麼。我們會讓他們在市場上看到。我覺得這很難。但我能說的是，從我們的角度來看，我們的能力仍在以指數級持續成長。

The opportunities we see continue to get bigger, and we're figuring out how to address, and deal with some of the implementation and deployment issues because it's not simple. I can't tell you that it's all in market today. It's stuff's complicated. If you can pull it off, it works really well.

我們看到的機會持續變得更大，而我們也在摸索如何因應並處理一些導入與部署的問題，因為這並不簡單。我不能告訴你說今天它已經全部在市場上了。事情很複雜。如果你能把它做成，它的效果會非常好。

The easiest way thing to do is generative AI, which is, I think last time I called it the, you might remember Keds sneakers. It's the run faster, jump higher strategy. It's good. It cuts out expenses. You can cut out call center people. All that's good. We're working on that. We do a bunch. It does millions of emails for us. People don't have to spend time doing it. It's all really good. I think the real benefit from this will come from agentic AI, where agents are talking to agents, and making decisions in subsecond real-time response rate that people then can't compete with you. We're building that.

最容易做的是生成式 AI；我想上次我把它比喻成——你可能記得 Keds 球鞋。這就是「跑得更快、跳得更高」的策略。它不錯。它能削減費用。你可以減少客服中心的人力。這些都很好。我們正在做這些。我們做了很多。它替我們處理了數百萬封電子郵件。人們不必花時間去做。這些都非常好。我認為真正的效益會來自代理式 AI，也就是代理與代理之間互相對話，並在次秒級的即時反應中做出決策，讓人們之後無法與你競爭。我們正在打造這個。

It's really complicated building that ecosystem. You got to get the right governance around it. You got to make sure you set up whatever metrics you give it will go get. You have to make sure measurement science is really important. We're working hard on that. We're excited about it. We think it offers a potential to really build off of what we did in transformative growth. We don't have the issue that some companies do. I don't know what our competitors' issues are, but I know other companies when I'm out talking to them have some issues in accessing legacy technology. We don't have that for many of our systems. We do for some, but not many of them, which gives us an ability to accelerate the agentic or ALLIE work.

打造那個生態系非常複雜。你必須建立正確的治理架構。你必須確保你給它的任何指標，它都會去追求。你必須確保衡量科學非常重要。我們正在努力做這件事。我們對此感到興奮。我們認為它有潛力在我們先前的轉型式成長成果之上再進一步擴展。我們沒有一些公司所面臨的問題。我不知道我們競爭對手的問題是什麼，但我在外面與其他公司交流時，我知道有些公司在存取既有（legacy）技術方面有一些問題。我們在許多系統上沒有這個問題。有些系統有，但不多，這讓我們有能力加速代理式或 ALLIE 的工作。

Got it. Okay, thank you.

了解。好的，謝謝。

Yaron Kinar, Mizuho.

Yaron Kinar，瑞穗（Mizuho）。

Thank you. Good morning. My first question is on the homeowners book. Why was the expense ratio up year-over-year? And would you still expect improvement for the full year?

謝謝。早安。我的第一個問題是關於住宅屋主保單（homeowners）這一塊。為什麼費用率（expense ratio）年增了？而你們是否仍預期全年會改善？

We reallocate expenses from time to time. There's slightly higher commissions related with bundling on that. While it looks expensive, it's good lifetime value. Let me put it that way because we've had the same question. I'm like, hey, wait a minute here. Like, where'd this point go? It relates to how we're driving value. We try to do it so it accurately reflects what each product gets and not just spread those costs by commission. We love the homeowners business. We think it's great. We think it's an underappreciated growth asset, not just given the market share numbers that Mario talked about.

我們會不時重新分攤費用。那裡有一些與綑綁銷售（bundling）相關、略高的佣金。雖然看起來費用較高，但它的客戶終身價值（lifetime value）很好。我就這麼說，因為我們也被問過同樣的問題。我會想，嘿，等一下。比如說，這個點數跑去哪了？這與我們如何創造價值有關。我們會盡量讓它能準確反映各產品應承擔的成本，而不是只按佣金把成本攤開。我們很喜歡住宅屋主保險業務。我們認為它很棒。我們認為它是一項被低估的成長資產，不只是因為 Mario 談到的市占率數字。

If you just think about severe weather, and you're looking for trends, people are going to need more insurance for their homes, the worse the weather it gets. We're really good at that business, so we like that business a lot. I think it has great potential.

如果你只看極端天氣，並在尋找趨勢，人們會需要為他們的房屋購買更多保險——天氣越糟就越需要。我們在這項業務上做得非常好，所以我們非常喜歡這項業務。我認為它有很大的潛力。

Just to clarify here. The reallocation of expenses, is that something that's going to flow through throughout the year? Do you still expect to see year-over-year improvement?

我想再釐清一下。費用的重新分攤，這會在全年持續反映出來嗎？你們仍預期會看到年對年改善嗎？

We don't do forecasts of expenses. To the extent we're spending the money to increase bundling, we like that. Yeah, it would be higher. We're still earning a great return. I wouldn't. Homeowners is a little less price sensitive than auto insurance, would be the other point I would just add to you as you're thinking this one through.

我們不做費用的預測。在我們投入資金以提高綑綁銷售的程度時，我們是樂見其成的。是的，那會更高。我們仍然能賺到很好的報酬率。我不會——另外我想補充的是，住宅屋主保險相較於汽車保險，對價格的敏感度稍低；你在思考這題時可以把這點納入。

Right. My second question, I realize it may still be relatively early, but we've had the closure in the Strait of Hormuz for what two months now. Do you expect gasoline prices and supply chain disruptions related to the closure to impact frequency and/or severity in both auto and home?

了解。我的第二個問題，我知道可能還算相對早期，但荷姆茲海峽（Strait of Hormuz）關閉到現在大概已經兩個月了。你們是否預期與關閉相關的汽油價格上漲與供應鏈中斷，會影響汽車與住宅的出險頻率（frequency）和／或損失嚴重度（severity）？

We don't know, would be the answer. I think I can give you some facts around it. About one third of driving is discretionary. About one third is for, like, going to work, and about one third is for, like, doing stuff you got to do, go to the grocery store or something like that. You're basically talking about one third of things people can decide they want to go on a shorter vacation or whatever. That obviously takes some time to factor in. If gas prices are $5, $6, people don't go as far. Maybe they share their car on riding to work. Maybe they don't go to the grocery store as often. There's various things that higher gas prices do result in fewer miles driven, which then lowers frequency.

答案是：我們不知道。我想我可以提供一些相關事實。大約三分之一的開車行為是可選擇性的。大約三分之一是為了，例如通勤上班；另外三分之一是為了你必須做的事，例如去雜貨店之類。你基本上談的是其中三分之一，人們可以決定要不要縮短假期或諸如此類。這顯然需要一些時間才會反映出來。如果油價是 5、6 美元，人們就不會開那麼遠。也許他們會共乘去上班。也許他們不會那麼常去雜貨店。油價上升會導致行駛里程下降的各種行為變化，進而降低出險頻率。

It's not a straight line. Like, you can't just say, oh, gas prices are $110, or it's $110 a barrel for oil, therefore we're going to have a half a point change in frequency. You just don't know. There's a million different variables to that. What we do know is we pay attention to frequency. We keep track of frequency. We do our claims. We do claim counts impact our reserving, and we get claim counts every day. To the extent they're changing, we're always looking at it. Then you have to decide, well, how long will it be there? Even when you have higher prices, you might get a temporary drop down, and then it goes back up.

但這不是一條直線關係。例如你不能只說，喔，油價是 110 美元——或原油每桶 110 美元——因此出險頻率就會變動 0.5 個百分點。你就是不知道。這裡面有上百萬個不同變數。我們知道的是，我們會關注出險頻率。我們會追蹤出險頻率。我們處理理賠。理賠件數會影響我們的準備金提列，而我們每天都會拿到理賠件數。只要它們在變化，我們就會一直盯著看。接著你還得判斷：這種情況會持續多久？即使價格較高，你可能會先看到短期下滑，然後又回升。

We track 50 million cars every day, every 15 seconds, so we know who's driving when.

我們每天追蹤5,000萬輛車、每15秒一次，因此我們知道誰在什麼時候開車。

That's on the frequency side. What about the severity side?

那是頻率面。那嚴重度面呢？

Severity, again, in general, higher petroleum prices roll through everything from plastic parts on cars to shingles. It has an upward impact on it. What happens to our costs? We don't see anything right now in severity, increasing severity of parts and stuff like that. There, it's a competitive market, so you just see what happens. We're not concerned about the price of oil and its impact right now on our profitability.

嚴重度方面，一般來說，較高的石油價格會影響從汽車塑膠零件到屋頂瀝青瓦等各種項目。它會帶來上行影響。那我們的成本會怎麼樣？目前我們在嚴重度上沒有看到任何變化，零件等項目的嚴重度並沒有上升。而且那是一個競爭激烈的市場，所以就看後續發展。就目前而言，我們不擔心油價及其對我們獲利能力的影響。

Thank you.

謝謝。

Paul Newsome, Piper Sandler.

Paul Newsome，Piper Sandler。

Good morning. Thanks for the call. Maybe a revisit to the competitive environment a little bit and talking about some of the states that have been not as attractive. Any thoughts about those states turning or some of the other states that were in between turning to be more positive environment or, but any color you could get I think would be helpful and interesting.

早安。謝謝這通電話。也許想再回頭談一下競爭環境，並聊聊一些一直不那麼有吸引力的州。對於那些州是否可能轉向，或其他介於中間的州是否會轉為更正向的環境，有沒有什麼想法？我覺得任何補充說明都會很有幫助也很有意思。

Well, it's Paul, it's really a question about the regulatory and operating environment, I think, rather than competitive is the way I'm hearing the question. Let me make sure I get it right before I answer.

嗯，Paul，我聽起來這更像是在問監管與營運環境，而不一定是競爭環境。在我回答之前，先確認我理解正確。

Well, I guess it's either one, right? If it's regulatory, then that's the thing to focus on. If it's competitive, then that's another thing. I guess.

嗯，我想兩者都算吧，對吧？如果是監管，那就聚焦在那個。如果是競爭，那就是另一個面向。我想是這樣。

Okay.

好的。

Investors need to hear more of the competitive piece than the regulatory piece at the moment.

目前投資人更需要聽到競爭面的內容，而不是監管面的內容。

Yeah, I'll go to both of them. Let's start with regulatory. Obviously, there were three large states we called out last year that we struggled to find a way in which we could earn an adequate return for our shareholders, give customers a good price, and grow, and so we didn't. Some of those are getting better. I'm really excited about what might happen in New York with Governor Hochul's doing. It will be a blow for freedom for insurance consumers to take the costs out of unnecessary, what I call fender bender litigation. That could be a huge benefit because New Yorkers pay a lot for insurance because there's a lot of these benefits being circulated.

好，我兩個都談。先從監管說起。顯然，去年我們點名了三個大型州，我們很難找到一種方式，既能為股東賺取足夠的報酬、又能給客戶好的價格、同時還能成長，所以我們沒有在那些州擴張。其中一些情況正在改善。我對紐約在Hochul州長推動的事情感到非常振奮。若能把我所稱的不必要、輕微擦撞事故訴訟的成本移除，這將是保險消費者自由的一大進展。這可能帶來巨大的好處，因為紐約人保費很高，部分原因就是這些給付在系統中被大量流轉。

Certainly, when people get hurt, their car gets wrecked, their bodies get bent up and stuff, they should be paid, and we're totally in favor of that, and that's what we do. That's what we like to do. Sometimes the system gets a little out of whack, and it needs to be course-corrected. We're thrilled about what they're planning to do or hoping to do in New York. If that happens, that would open a giant growth market for us. We have a big share in New York, particularly in the five boroughs. We've been really strong there for a long time. We have a great agency force. It's got tight media markets, so our direct operations work really well there. We have good independent agent relationships.

當然，當人受傷、車子被撞毀、身體受創等等，他們應該獲得賠付，我們完全支持，這也是我們在做的事。這正是我們想做的。但有時制度會有點失衡，需要校正。我們對他們計畫或希望在紐約推動的改革感到非常興奮。如果真的發生，對我們而言將開啟一個巨大的成長市場。我們在紐約市，特別是五大行政區，市占率很高。我們在那裡長期以來一直很強。我們有很棒的代理人隊伍。那裡媒體市場集中，所以我們的直營運作在那裡非常有效。我們也有良好的獨立代理人合作關係。

That would be a great place for us to grow, and we hope that they can do that because it'll be good for our customers and consumers in general. If I just go up to the competitive environment, it continues to be highly competitive in auto insurance as Mario talked about.

那會是我們很適合成長的地方，我們也希望他們能做到，因為這對我們的客戶以及整體消費者都有利。如果我轉到競爭環境，如Mario所說，汽車保險仍然競爭非常激烈。

The top five continue to battle it out. You see some of the, they're not small, but they're not in the top five. Some of the independent agent carriers have had volumes go down. Particularly two big independent agent commercially focused companies have lost some share there. We feel good about our competition in auto insurance against the top 5.

前五大業者仍在激烈競爭。你也看到一些公司——它們不算小，但不在前五大。一些獨立代理通路的承保公司保單量下滑。特別是兩家以商業保險為主、透過獨立代理的較大型公司，在那裡流失了一些市占。我們對於在汽車保險領域與前五大競爭的態勢感到有信心。

Where we're really starting to pick up some momentum against competition is in the homeowners business. Mario talked about that, 81% of the country. Some of those top five either don't really sell their own product and have underwriting margin to work on in that space, or haven't had as good results as they would like. They're being less aggressive in that space. We think there's great potential to grow in the homeowners business given that competitive set. Anything Mario or anybody would add to that?

我們真正開始相對競爭對手累積動能的，是住宅險業務。Mario提到過，涵蓋全國81%的地區。前五大中的一些業者，要嘛並不真正銷售自家產品、在該領域可運用的承保利潤空間有限；要嘛業績結果不如預期。因此他們在那個領域的進取程度降低。基於那樣的競爭組合，我們認為住宅險業務有很大的成長潛力。Mario或其他人要補充嗎？

No, I think, I think you nailed it, Tom. The, the only thing I'd say is when you look at it's a highly competitive market, as Tom said. I think when you look at our results and we continue to generate new business at historically high levels, it's across distribution channels. We're leveraging all the capabilities Tom talked about early on, and we're competing effectively both in the auto and the homeowner space. We like our chances to be able to continue to do that going forward.

沒有，我覺得你講得很到位，Tom。唯一要補充的是，如Tom所說，這是一個高度競爭的市場。但你看我們的成果，我們持續以歷史高檔的水準創造新業務，而且是跨各個通路。我們正在運用Tom先前提到的各項能力，並且在汽車險與住宅險兩個領域都能有效競爭。我們看好自己未來持續做到這一點的機會。

That's great. As a second question, maybe turning to the home business, it covered a lot of little companies that are talking about the business moving on the margin to more excess and surplus lines for home insurance. Any thoughts about that trend, if you think it's just kind of a temporary thing or it's a part of what matters for you? I would imagine given your very middle-of-the-road product for home insurance, it's not a huge piece, but just curious.

很好。第二個問題，回到住宅險業務，很多小型公司都在談住宅保險在邊際上更多轉向超額與剩餘保險（E&S）。你們對這個趨勢有什麼看法？你們覺得這只是暫時現象，還是對你們而言是重要的一部分？我猜以你們相對中間定位的住宅險產品來看，這應該不是很大的部分，但我只是好奇。

Not a huge piece of excess and surplus lines for us or for our competitors?

你是指對我們或對競爭對手而言，超額與剩餘保險（E&S）都不是很大的一塊嗎？

Yes. I would imagine it's pretty small for you folks.

是的。我想對你們來說應該很小。

Yeah. We have an excess and surplus lines business. It's the Northlight. It's grown reasonably well. Just to help educate everybody else who's not as in deep, Paul, as you are on. Excess and surplus lines are where there's not enough availability in a market, and a customer goes out to, like, 2 or 3 companies, can't get an offer, and so then, somebody can offer them an excess and surplus lines company, which is, I'm going to call it lightly regulated as it relates to price, as opposed to tightly regulated in homeowners. We have a company that does that.

是的。我們有超額與剩餘保險（E&S）業務。那是Northlight。它成長得相當不錯。也稍微幫其他沒有像你這麼深入了解的人補充一下，Paul。所謂超額與剩餘保險，是指某個市場的承保供給不足，客戶去找例如2到3家公司都拿不到報價，於是就有人可以透過超額與剩餘保險公司提供承保；相較於住宅險在價格上屬於嚴格監管，我會把E&S稱為在價格方面「相對寬鬆監管」。我們有一家公司在做這件事。

We prefer to do it in the regular lines. If we can't sell it in the regular lines, we don't necessarily use our excess and surplus lines because we don't like the market. Like, occasionally they might, we might use excess and surplus lines for a really well-priced risk. In general, if we don't like the state for homeowners, we probably don't like it for excess and surplus lines either. We do want it so that we can be available for customers that have it.

我們偏好在一般承保線（regular lines）來做。如果我們無法在一般承保線把它賣出去，我們不一定會改用超額與剩餘承保線（excess and surplus lines），因為我們不喜歡那個市場。例如，偶爾在風險定價非常划算的情況下，我們可能會使用超額與剩餘承保線。一般而言，如果我們不喜歡某個州的住宅險市場，我們大概也不會喜歡該州的超額與剩餘承保線。但我們確實希望具備這個選項，讓我們能為有此需求的客戶提供服務。

On top of that we're probably the biggest broker of homeowners insurance in the country, because to serve our Allstate agent customers well, when we can't offer a product in Florida or California or something like that, we have arrangements with other companies that we can sell their product for. That's again a number with a B on it in terms of how much product we sell that way.

除此之外，我們可能是全國最大的住宅險經紀商，因為為了更好地服務我們的 Allstate 代理人客戶，當我們無法在佛羅里達或加州之類的地方提供產品時，我們與其他公司有合作安排，可以代銷他們的產品。同樣地，就我們透過這種方式銷售的產品規模而言，那也是一個以「十億」（B）為單位的數字。

Good. Appreciate it. Thank you very much.

很好。感謝。非常謝謝。

Tracy Benguigui, Wolfe Research.

Tracy Benguigui，Wolfe Research。

Thank you. Good morning. You started the earnings call by giving a demo on your ad campaign. How should we think about ad spend budget this year versus last, and any expense ratio impacts and PIF growth prospects as a result?

謝謝。早安。你在財報電話會議一開始展示了你們的廣告活動示範。我們應該如何看待今年相較去年廣告支出預算的變化，以及因此對費用率的影響與在保單件數（PIF）成長前景？

It's a highly competitive market. We've dialed up advertising significantly over the last four years. We dialed that up with increased sophistication, so there's upper and lower funnel. Upper funnel being the stuff you saw, lower funnel being very specific. We find -- Chris is shopping for insurance, and we, like, give her an ad at that moment on her addressable TV or on the web or something like that. There's upper and lower funnel. We've increased our lower funnel advertising this year, which has better metrics. It's easier to do metrics on it. It's like run ad on the Super Bowl who's watching it? Do they buy anything from you?

這是一個競爭非常激烈的市場。過去四年我們大幅提高了廣告投放。我們也以更高的成熟度來提升投放，因為有上層漏斗與下層漏斗之分。上層漏斗就是你看到的那些內容；下層漏斗則非常精準。我們會發現——例如 Chris 正在比價買保險時——我們就在那個當下，透過可定向的電視（addressable TV）或網路之類的渠道投放廣告給她。所以有上層與下層漏斗。今年我們增加了下層漏斗的廣告投放，因為它的指標表現更好。也更容易衡量。像是在超級盃投放廣告，誰在看？他們有沒有跟你買任何東西？

It's not as easy to find out whether that's economic as when. We've shifted more to lower funnel. We spend relative to where our economics are. We have economic measures, but we don't spend all the way up. Like recently, we were looking at should we spend more? Sometimes you just want to make sure the system works really hard. You don't want advertising to be the only thing you do to drive growth, because you end up in a systems theory where we spend more, Progressive spends more, so leads go up in cost, so we spend more, so they spend more. You have to be careful that you don't feed a beast you don't want to feed. We're highly precise, I guess I would say, and disciplined about it.

要判斷那樣是否具經濟效益，並不像（下層漏斗那種）那麼容易。我們已把更多比重轉向下層漏斗。我們的支出會依據經濟性來決定。我們有經濟性衡量指標，但不會一路加碼到極限。例如最近我們在看是否應該花更多。有時你只是想確保整個系統運作得非常努力。你不希望廣告成為推動成長的唯一手段，因為最後會落入一種系統動力學：我們花更多、Progressive 也花更多，導致名單（leads）成本上升，所以我們再花更多、他們也再花更多。你必須小心，不要餵養一個你不想餵養的怪獸。我想我們做得非常精準，而且很有紀律。

That said, when we think we can advertise and if we think we could spend more money and grow more and get a great combined ratio, we will. Right now, we like our economics. Our economics are better this year than they were last year. Some of that's just getting better at executing. It isn't that the market's really changed some. We've just gotten better at our close rates.

話雖如此，當我們認為可以透過廣告投放、如果我們覺得多花一些錢能帶來更多成長，並且仍能取得很好的綜合率（combined ratio），我們就會這麼做。目前我們喜歡我們的經濟性。今年的經濟性比去年更好。其中一部分只是我們執行得更好。並不是市場真的有很大的改變。我們只是把成交／轉換率（close rates）做得更好了。

Excellent. Shifting gears, can we talk about asset allocation? You doubled your equity holdings since September, so it's about 12% of your total portfolio. What is that relative to your equities asset allocation target?

很好。換個話題，我們可以談談資產配置嗎？你們自九月以來把股票持有部位加倍，現在約占整體投資組合的 12%。這相對於你們股票資產配置目標而言是什麼位置？

I'll let John answer that, since that's also part of what he does besides being our Chief Financial Officer. We try to make sure everybody knows at least 2.5 jobs. I would just say, which he probably wouldn't want to say to himself, that we're really good investments. We manage it around. We're proactive. We think about it from an enterprise standpoint. You can see the numbers on the charts. We have good confidence that we can generate good returns on capital, and you see it flow through our P&L, particularly this quarter.

我讓 John 來回答，因為除了擔任我們的財務長之外，這也是他工作的一部分。我們會確保每個人至少知道 2.5 份工作。我只想說一句——他可能不會自己這麼說——我們的投資做得非常好。我們會圍繞它來管理。我們很積極主動。我們會從企業整體的角度來思考。你可以在圖表上看到那些數字。我們很有信心能創造良好的資本報酬，你也能看到它反映在我們的損益表上，尤其是本季。

Yeah. Hey, thanks for the question, Tracy. The way I think about it is I take a step back and really, go back to the presentation, think about how we think about capital allocation in general. We have a lot of different things we can do. We think about the overall enterprise context as we do it. We also think about what's going on in the market environment at any part in time.

是的。Tracy，謝謝你的問題。我的思考方式是先退一步，回到簡報內容，從我們整體如何看待資本配置來思考。我們有很多不同的事情可以做。我們在執行時會考量整體企業情境。我們也會考量在任何時間點市場環境正在發生什麼。

You've seen us in our portfolio change our allocation probably more than most of our peers, whether that's equity or whether that's fixed income changing our exposure to rate via duration and the rest. Because we're active, I don't know that I could point you to a specific asset allocation target.

你們也看到，我們在投資組合中調整配置的幅度可能比多數同業更大，不論是股票，或是固定收益——例如透過存續期間（duration）等方式調整對利率的曝險。因為我們是主動操作，我不確定我能指給你一個特定的資產配置目標。

There's a range that's defined by past behaviors that probably gives you a pretty good idea of what we're likely to operate within. When we do put more money to work, particularly in equities, we try and take a mid to longer range view on it. We are economic investors. We're not just trying to manage to a yield target at any point in time. We think by delivering economic value that does accrue to increase net investment over time, and it's a more cerebral way of going about it.

不過有一個區間，是由過去的行為所界定，應該能讓你相當清楚我們大概會在什麼範圍內運作。當我們投入更多資金、特別是在股票上時，我們會盡量採取中長期的視角。我們是以經濟性為導向的投資人。我們不是只想在任何時間點去管理一個殖利率目標。我們認為透過創造經濟價值，會隨時間累積並提升淨投資收益（net investment），這是一種更偏理性、深思熟慮的做法。

We're not necessarily trying to measure that quarter by quarter, and we're taking a longer look. The amount that we put to work recently has that in mind. If you look back, say, six months ago, the environment was a little less certain. We had a number of things going on. We have a little bit more clarity and felt good about putting money to work, and we'll see how it turns out in the coming quarters and years.

我們不一定會逐季去衡量它，而是用更長的時間視角來看。我們最近投入的資金規模也是基於這樣的想法。如果你回頭看，例如六個月前，環境的不確定性更高一些。當時有不少事情在發生。現在我們更清楚一些，也覺得把資金投入運作是合適的；接下來幾季與未來幾年就看看結果如何。

Okay. It sounds like your approach is more dynamic than static, so could we foreseeably see that percentage growing if you like that asset class?

了解。聽起來你們的做法比較偏動態而非靜態，所以如果你們喜歡那個資產類別，我們是否可以預期那個比例還可能上升？

I would say that it's dynamic, but it's well-governed, and you could probably gauge most of the range of our future activities by the way that we've conducted ourselves in the past.

我會說它是動態的，但治理完善；而且你大概可以從我們過去的作法，推估我們未來大部分活動的區間。

Thank you.

謝謝。

Yeah. We're not likely to have an 80% equity allocation.

是的。我們不太可能把股票配置提高到 80%。

Okay, got it.

好的，了解了。

Pablo Singzon, JPMorgan.

Pablo Singzon，摩根大通。

Hi. Thank you for squeezing me in. Just one for me. I wanted to shift to AI again, but this time as it relates to your distribution strategy and how you reach customers. I presume it helps direct distribution, but how do you think it affects your agents, whether captive or independent? There's an argument that it makes them productive, but do you think AI ultimately shifts business away from them? Thank you.

您好。謝謝讓我插個問題。我只有一個問題。我想再回到 AI，但這次是就你們的通路／分銷策略以及你們如何觸達客戶而言。我推測它有助於直銷，但你們認為它會如何影響你們的代理人，不論是專屬（captive）或獨立代理人？有人認為它會提升他們的生產力，但你們覺得 AI 最終會把業務從他們那裡轉移走嗎？謝謝。

Sorry. What makes the agents more productive?

不好意思。是什麼讓代理人更有生產力？

Just the use of AI. Yep. That's sort of like the argument today.

就是使用 AI。對。這大概就是目前的論點。

Oh, AI.

喔，AI。

Right. Yep.

對。是的。

Sorry. I missed the two probably most talked about letters in these days. AI can help them in a whole bunch of ways. First, it can help us have a better product and better pricing and deliver better service for people. That's in general, just it. It'll help us be a better company. Secondly, as it relates to their specific work, we think it will remove a lot of service work out of agents' offices. Things they had to do before, they won't have to do anymore. We're actively working to get that work out of their offices. Secondly, it will help them be smarter and on behalf of agents who provide more advice and do less individual work.

抱歉。我漏聽了這幾天可能最常被提到的兩個字母。AI 可以在很多方面幫助他們。首先，它可以幫助我們提供更好的產品、更好的定價，並為客戶提供更好的服務。總體而言，就是這樣。它會讓我們成為一家更好的公司。第二，就他們的具體工作而言，我們認為它會把大量的服務性工作從代理人的辦公室移除。以前他們必須做的事情，之後就不必再做了。我們正積極把那些工作從他們的辦公室移出去。另外，它也會讓他們更聰明，代表那些提供更多建議、做更少個別作業的代理人。

Let's say we were going to do an insurance review for you. An agent might have to go pull your records, see what you've got, see what, how old your kids are, figure out. With both advanced computing, whether you want to call it machine-based learning, AI, whatever, we can help them do that work ahead of time, so they're really delivering the work, and it's like they have an analyst working for them to help them. The other thing that AI can do is really in the moment. We have in market today, something called Customer Engagement Sidekick, that helps you really do a better job of engaging with customers.

假設我們要為您做一次保險檢視。代理人可能得去調出您的資料，看看您目前有哪些保障、您的孩子多大了等等，然後再做判斷。透過先進運算，不管你想稱之為機器學習、AI 或其他，我們可以幫他們事先把這些工作做好，讓他們真正把成果交付出來；就像有一位分析師在替他們工作、協助他們一樣。AI 另一個能做的事情是即時支援。我們目前在市場上有一個叫做 Customer Engagement Sidekick 的工具，能幫助你更有效地與客戶互動。

Because you might have, if you're doing 50 calls a day or something like that it's always good to have somebody, hey, this is what I'm kinda hearing. Maybe you should go here. Here's the tonality we're talking about. We think it'll help them do a much better job for those people who want somebody in between them. Their AI can also just sell directly, and we're live in the market doing that right now on a particular product. It's more of a learning, but it's doing it in three states. It's closing policies. We're just seeing what we learn from that. You just have to be there to meet the customers. I think it will help those agents who have good relationships with people improve their relationships.

因為如果你一天要打 50 通電話之類的，總是很需要有人在旁邊提醒：嘿，我大概聽到的是這個。也許你應該往這個方向談。我們現在談話的語氣／情緒是這樣。我們認為，對於那些希望有人在中間協助的人，這會幫助代理人把服務做得更好。AI 也可以直接銷售，而我們現在就已經在市場上針對某個特定產品上線這樣做。這比較像是在學習，但目前在三個州進行。它正在完成保單成交。我們只是看看能從中學到什麼。你必須在那裡，才能接觸到客戶。我認為它會幫助那些與客戶關係良好的代理人，進一步改善他們的關係。

It'll help other agents build more relationships. Those people who just don't feel like dealing with it and would just as soon deal with a computer, we'll be there for them too. Thank you. Is that the last question? Okay. Thank you all for -- we obviously had a great quarter. We had strong earnings, increased growth with transformative growth. We think it's showing up. We went through the market share gains. We look forward to your engagement with Allstate, and we'll keep working on creating more shareholder value. Thank you.

它也會幫助其他代理人建立更多關係。至於那些不想處理、寧可跟電腦打交道的人，我們也會為他們提供服務。謝謝。這是最後一個問題嗎？好的。謝謝各位——我們顯然有一個非常好的季度。我們的獲利強勁，成長加速，且是具變革性的成長。我們認為成果正在顯現。我們也談到了市占率的提升。我們期待各位與 Allstate 的持續互動，我們也會繼續努力創造更多股東價值。謝謝。

Thank you, ladies and gentlemen, for your participation in today's conference. This does conclude the program. You may now disconnect. Good day.

各位女士、先生，感謝您參與今天的電話會議。本次會議到此結束。您現在可以掛線。祝您有美好的一天。