FIS
FIS 5Price and target zones价格与目标区间
Near the buy zone接近买入区间
What the company does公司做什么
FIS sells the software that banks, credit unions, and capital-markets firms use to run core banking, digital banking, payments, and trading operations. In 2024, FIS sold most of its Worldpay merchant-payments business to focus on these two core areas. In January 2026, FIS sold its remaining Worldpay stake and used the proceeds to buy Global Payments' Issuer Solutions business.FIS（富达国民信息服务公司）向银行、信用合作社和资本市场机构销售软件，用于运行核心银行业务、数字银行、支付处理和交易系统。2024年，FIS出售了其Worldpay商户支付业务的大部分股权，以专注于这两大核心业务。2026年1月，FIS出售了其剩余的Worldpay股权，并用所得资金收购了Global Payments的发卡方解决方案业务。
Earns from营收来源 FIS earns money mainly from recurring software licensing, processing, and subscription fees under long-term contracts with banks and financial firms.FIS的主要收入来自与银行和金融机构签订的长期合约中的经常性软件许可费、处理费和订阅费。
Revenue breakdown营收构成 share of revenue营收占比
This segment supplies the core processing systems banks use to manage accounts, loans, and payments, plus digital banking tools and payment technology. Customers range from large global banks to small community banks and credit unions, in more than 100 countries.该部门提供银行用于管理账户、贷款和支付的核心处理系统，以及数字银行工具和支付技术。客户范围从大型全球银行到小型社区银行和信用合作社，遍及100多个国家。
This segment sells software that helps capital-markets firms handle trading, treasury management, risk, and back-office securities processing. It serves both buy-side firms, such as asset managers, and sell-side firms, such as investment banks.该部门销售软件，帮助资本市场机构处理交易、资金管理、风险控制和后台证券处理业务。服务对象既包括买方机构（如资产管理公司），也包括卖方机构（如投资银行）。
Investment thesis投资逻辑 from the watchlist来自关注列表
What's working
- The best fact in the 8/04 print: FCF guidance was RAISED to $2.15-2.25B (+33-39%) while EPS guidance was CUT — cash is arriving faster than the P&L, and cash is what services $20.3B of net debt. Q2 FCF tripled YoY to $525M; equity FCF yield ~9.9%; dividend covered at 41% of FCF. Banking +6.1% pro forma at a 45.8% EBITDA margin (+179bp), adj EBITDA margin +193bp, ~80% recurring revenue on the highest switching costs in enterprise software. CEO Stephanie Ferris bought 19,846sh at $50.39 = $1,000,040 on 3/05 — a real open-market purchase, now ~15% underwater. (Director Goldstein's ~$57-60k buys every Jan/Apr/Jul/Oct 15th are a scheduled fee-deferral, not conviction — discount them per the RSU pitfall.)
Red flags
- The headline 6.6x P/E is fiction and the error is 3x. Net margin 28% against op margin 21% fired
pattern-net-margin-above-operating-margin-is-a-tripwire: Q1 2026 shows $259M of pretax income and a $2,214M "earnings from equity interest" line — the gain on selling the remaining 45% Worldpay stake to GPN, closed 1/9/26. Rebuilding TTM quarter-by-quarter ($0.50 + $0.98 + $4.58 + $0.45 = $6.51, matching Yahoo exactly) strips to core TTM EPS ~$2.22 → true trailing P/E 19.3x. The vendor's 6.38x forward is also wrong-year — on the company's own FY26 guide of $6.15-6.24 adjusted it is 6.9x - The load-bearing risk: the deleveraging arithmetic does not close on cash flow. 3.6x → 2.8x on ~$5.6B EBITDA needs $4.6B of paydown against $2.2B FCF − $0.91B dividends = $1.29B/yr — 2.3 years at best, 3.6 at flat EBITDA, against a promised end-2027. That is why Capital Markets is being shopped, and it makes the 2027 buyback restart conditional on an asset sale the buyer knows they need → [[pattern-deleveraging-target-that-needs-an-asset-sale]]
- ROIC 3.7% → 4.9%, below WACC — $43B for Worldpay in 2019, ~$24B written off, retained earnings −$20.6B, and then $7.7B of fresh debt in Jan 2026 to buy Issuer Solutions (legacy TSYS) from a seller that wanted it gone. Debt/assets 39.8% → 48.6% in one quarter; goodwill+intangibles 79% of assets; tangible book −$18.3B (−$35.35/sh); $4.15B of commercial paper
- Moat decays by attrition, not assault — nobody rips out an incumbent core, but Thought Machine/Mambu/10x Banking (10M+ live accounts) take every new charter and every modernization budget. 5.3% organic growth is what that looks like early. AI revenue is explicitly not in FY26 guidance — optionality, value it at zero
Watch
- ANALYZED 8/06 — first file ever. WATCH — no position. FV $42-54 (mid ~$48) — spot sits at the BOTTOM EDGE of fair value, not below it.
- The 8/04 cut: FY26 revenue $13.63-13.70B and adj EPS $6.15-6.24 (both below the $13.8B / $6.28 consensus), Q3 $1.58-1.62 vs $1.65, and Capital Markets growth cut 225bp to 3.0-3.5%. Ferris: "We are not satisfied with our performance in Capital Markets." Half the cause was a self-admitted internal execution miss on professional-services conversion, not just the known UBS/Credit Suisse attrition — a demand problem you wait out, an execution problem you have to believe gets fixed by the people who caused it. Every covering firm cut targets on 8/05 (UBS Buy → Neutral, $63 → $49; WFC $67→$58; JPM was $80 in Feb); mean $56.55 and still falling
- Two models are void, and saying so is the finding: Graham prints $67.28 off the fake $6.51 EPS and a book that is 100%+ goodwill; naive DYT prints $73.64 but fails 3 of 4 guard questions — the yield rose on a −41% price, the dividend was CUT 31% in 2024 ($0.52→$0.36/qtr, now rebuilt to $0.44), and the 2.39% 5yr band was set when FIS traded $90-150 as a "payments compounder." Reset DYT: $39-50. DDM $38-48; 11-13x FCF $47-56
- No per-share tailwind for ~two years — buybacks paused until 2.8x, after retiring 15% of shares in four years. Entry $36-42, strong <$38 (11.6% FCF yield, 4.6% covered dividend). Trim 8x fwd (deliberately 8x not 9x: the site derives the dollar from Yahoo's forward P/E, whose implied $6.71 EPS runs ~8% above the company's own $6.20 guide, so 8x lands at the ~$54 ceiling). ⏳ Gated on the Q3 print, early Nov.
- Sleeve placement is deliberate — 4.11% fails Yield-Today's ≥5% "coupon does the work" test, and this is a multiple with a dated catalyst. It sits beside FISV and GPN because it IS them.
Would upgrade on
- Upgrade: price $36-40 Q3 confirming Capital Markets stabilised a Capital Markets sale that visibly closes the deleveraging gap cluster insider buying below $42.
👀 🔴 8/06 分析完成 —— 史上第一份报告。 观察 —— 无仓位。 合理价值 $42-54（中值约 $48）—— 现价位于合理价值的下沿，而不是低于它。 ** 🚩 6.6x 的表面市盈率是假的，误差达 3 倍。 ** 净利率 28% 高于营业利润率 21%，触发了 pattern-net-margin-above-operating-margin-is-a-tripwire：2026 年 Q1 显示 税前利润 $259M，而“权益法投资收益”一行为 $2,214M —— 这是把剩余 45% 的 Worldpay 股权出售给 GPN 的收益，交易于 1/9/26 完成。 逐季重建 TTM（$0.50 + $0.98 + $4.58 + $0.45 = $6.51，与 Yahoo 完全吻合）后，剥离出 核心 TTM 每股收益约 $2.22 → 真实的历史市盈率 19.3x。 供应商给出的 6.38x 前瞻市盈率也用错了财年 —— 按公司自己的 FY26 指引 调整后每股收益 $6.15-6.24 计算，是 6.9x。 ✅ 8/04 财报中最好的一点：自由现金流指引被上调至 $2.15-2.25B（+33-39%），而每股收益指引被下调 —— 现金到账的速度快于利润表所显示的，而现金正是偿付 $20.3B 净负债所需要的。 Q2 自由现金流同比增至三倍，达 $525M；股权自由现金流收益率 约 9.9%；股息由 41% 的自由现金流覆盖。 银行业务 同口径 +6.1%，EBITDA 利润率 45.8%（+179bp），调整后 EBITDA 利润率 +193bp，约 80% 为经常性收入，建立在企业软件中最高的转换成本之上。 CEO Stephanie Ferris 于 3/05 以 $50.39 买入 19,846 股 = $1,000,040 —— 一笔真实的公开市场买入，目前浮亏约 15%。 （董事 Goldstein 在每年 1/4/7/10 月 15 日约 $57-60k 的买入是既定的薪酬递延安排，不是信心信号 —— 按 RSU 陷阱笔记应予折扣。 ） 🔴 8/04 的下调：FY26 收入 $13.63-13.70B、调整后每股收益 $6.15-6.24（均低于 $13.8B / $6.28 的市场预期），Q3 为 $1.58-1.62 对预期 $1.65，资本市场业务增速下调 225bp 至 3.0-3.5%。 ** Ferris：“我们对资本市场业务的表现并不满意。 ” 其中一半原因是公司自己承认的专业服务转化环节的内部执行失误，而不只是已知的 UBS/瑞信合并造成的客户流失 —— 需求问题可以等，执行问题则要相信造成它的这批人能修好它。 8/05 每一家覆盖该股的机构都下调了目标价（UBS 买入 → 中性，$63 → $49；WFC $67→$58；JPM 在 2 月还是 $80）；均值 $56.55，且仍在下行。 🚩 最关键的风险：去杠杆的算术在现金流上算不平。 ** 以约 $5.6B EBITDA 计，3.6x → 2.8x 需要 偿还 $4.6B 债务，而可用资金只有 $2.2B 自由现金流 − $0.91B 股息 = $1.29B/年 —— 最快 2.3 年，EBITDA 不增长则要 3.6 年，而承诺的时点是 2027 年底。 这正是资本市场业务被摆上货架的原因，也让 2027 年重启回购变成了一笔买方明知卖方非卖不可的资产出售的附带条件 → [[pattern-deleveraging-target-that-needs-an-asset-sale]]。 🚩 ROIC 3.7% → 4.9%，低于加权平均资本成本 —— 2019 年花 $43B 收购 Worldpay，减记约 $24B，留存收益 −$20.6B，然后在 2026 年 1 月又新增 $7.7B 债务，从一个巴不得脱手的卖方手中买下 Issuer Solutions（原 TSYS）。 负债/资产 一个季度内由 39.8% 升至 48.6%；商誉+无形资产占资产 79%；有形账面价值 −$18.3B（每股 −$35.35）；商业票据 $4.15B。 🚩 护城河是被蚕食的，不是被强攻的 —— 没人会拆掉在用的核心系统，但 Thought Machine/Mambu/10x Banking（活跃账户超 1,000 万）拿走了每一张新牌照和每一笔现代化改造预算。 5.3% 的有机增长正是这一过程的早期形态。 AI 收入 明确不包含在 FY26 指引内 —— 属于期权价值，现在应按零计。 ❌ 有两个模型失效，而指出这一点本身就是结论： Graham 模型基于虚假的 $6.51 每股收益和 100% 以上由商誉构成的账面价值，得出 $67.28；简单的股息收益率理论得出 $73.64，但 4 项检验中有 3 项不通过 —— 收益率上升是因为股价 −41%，股息在 2024 年被削减 31%（$0.52→$0.36/季，现已重建至 $0.44），而 2.39% 的 5 年区间是 FIS 以“支付行业复利机器”身份交易在 $90-150 时形成的。 重置后的股息收益率理论：$39-50。 ** 股利折现模型 $38-48；11-13x 自由现金流对应 $47-56。 ⚠️ 未来约两年没有每股层面的顺风 —— 回购暂停至杠杆降到 2.8x，此前四年已回购注销 15% 的股份。 买入区间 $36-42，$38 以下为强力买入（11.6% 自由现金流收益率，4.6% 且有覆盖的股息）。 减持 8x 前瞻市盈率 （刻意用 8x 而非 9x：网站是用 Yahoo 的前瞻市盈率反推美元价位的，而其隐含的 $6.71 每股收益比公司自己 $6.20 的指引高约 8%，因此 8x 恰好落在约 $54 的上限）。 ⏳ 等待 11 月初的 Q3 财报。 ** ⚠️ 归入哪个板块是刻意选择的 —— 4.11% 达不到“当期收益”板块 ≥5%“靠票息赚钱”的标准，而这笔投资买的是一个有明确催化剂日期的估值倍数。 它和 FISV、GPN 放在一起，因为它就是它们。 ** A2026-08-06**
Competitors竞争对手
Why customers stay客户为什么留下
FIS's banking and capital-markets software is built deep into customers' daily operations, so a switch to a rival provider is costly and slow. This makes customer relationships sticky and gives FIS predictable, recurring revenue. Its scale creates cross-sell chances that smaller rivals cannot match, though FIS must keep modernizing its older technology to compete with newer, cloud-native rivals.FIS的银行和资本市场软件深深嵌入客户的日常运营中，因此更换供应商既费钱又费时。这使客户关系具有很强的粘性，并为FIS带来可预测的经常性收入。凭借规模优势，FIS拥有较小竞争对手无法比拟的交叉销售机会，但公司仍须不断更新老旧技术，以应对更新、云原生的竞争对手。
Related companies相关公司 🔴 Bank-processor de-rating (new 8/06)
These are not three ideas — they are one idea held three ways: levered US bank/payment-processing incumbent de-rated on a guidance cut intact switching-cost moat management-credibility problem ~6x forward adjusted earnings.
What separates them: FIS pays a 4.11% dividend covered at 41% of FCF but carries 3.6x leverage with buybacks paused to 2027 and a deleveraging plan that needs an asset sale; FISV pays nothing but has…