After the North American internet pullback, whether AI fund flows can shift to the internet sector through EPS has become the core question

Overview

This report is Morgan Stanley's update on the trading position and valuation status of the North American internet sector, focusing on whether AI fund flows will shift into the internet sector through EPS realization. The report uses peer comparison tables, short-term stock price performance, EV/Revenue, EV/GP, EV/EBITDA, P/E, FCF yield, EPS, and short interest ratios to compare sub-sectors such as digital advertising, e-commerce, travel, shared economy, gaming, and real estate technology.

Core views

The core view is that the North American internet sector has pulled back meaningfully recently, but the industry view remains Attractive. Digital advertising and internet leaders are under pressure, especially GOOGL, AMZN, and META, which have underperformed the market; meanwhile, travel and the shared economy have been stronger. On valuation, some earnings multiples trade at discounts to historical averages, suggesting that if AI investment can convert into EPS growth, the internet sector may regain fund inflows; however, sales multiples remain above historical averages, and cash treatment of SBC would also materially raise valuation multiples for some companies.

Analysis framework

The report uses horizontal peer comparison and deviation-from-historical-average analysis: it first examines one-week price performance and market-cap-weighted group performance, then compares major internet companies' 2026E/2027E revenue, gross profit, EBITDA, EPS, FCF yield, and P/E multiples, and finally assesses valuation quality through NTM EV/EBITDA, NTM EV/Sales, and SBC cash-adjusted valuation.

Methodology notes

• Valuation ComparisonPeer Comp Sheet

  Compare EV/Revenue, EV/GP, EV/EBITDA, P/E, and FCF yield by sub-sector

  By grouping digital media, e-commerce, travel, shared economy, gaming, and real estate technology, the report assesses the relative valuation positioning of companies and sub-sectors.

• Historical Average ComparisonNTM Multiples vs. 5-year and 10-year Averages

  Compare current NTM EV/EBITDA and NTM EV/Sales with historical averages

  The report notes that NTM EV/EBITDA is about 10%/16% below its 5-year/10-year averages, while NTM EV/Sales is about 13%/16% above its 5-year/10-year averages, indicating that earnings multiples and sales multiples do not send fully consistent valuation signals.

• Earnings Quality AdjustmentSBC Cash Treatment

  Recalculate EV/EBITDA after treating stock-based compensation (SBC) as a cash cost

  This treatment increases EV/EBITDA multiples for some companies; the report shows an average increase of about 22% for gaming and about 44% for travel/shared economy/real estate technology.

• Price Performance AnalysisOne-week Market-cap-weighted Performance

  Compare short-term stock price performance of digital advertising, e-commerce, travel, and shared economy on a market-cap-weighted basis

  This method is used to identify short-term fund preferences: digital advertising and e-commerce are weaker, while travel and the shared economy are relatively stronger.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• North American Internet Sector

  The main coverage target of the report, with an Attractive industry view.

  Strengths

  Broad company coverage, many sub-sectors, some earnings multiples below historical averages, and AI-related EPS realization could bring fund inflows back.

  Weaknesses

  The sector fell about 6% over the past week, sales multiples remain above historical averages, and short-term risk appetite is weak.

  Comparison

  Relative to SPX/NDX performance of -2%/-5% over the same period, the overall internet sector was weaker; however, travel and the shared economy significantly outperformed internally.

  Risks

  AI investment may fail to convert into EPS, valuations may continue to compress, and SBC and regulatory pressure may affect earnings quality.

• AMZN/GOOGL/META

  Large internet and AI-related core names, serving as important reference points in the report's discussion of fund flows and EPS rotation.

  Strengths

  2026E EPS multiples trade at discounts to TTM averages; if AI commercialization improves profits, valuation recovery conditions are in place.

  Weaknesses

  Last week AMZN and META were about -5%, while GOOGL was about -8%, showing clear short-term capital pressure.

  Comparison

  The report gives 2026E EPS multiples of about 25x for AMZN, 23x for GOOGL, and 17x for META, roughly 22%/11%/20% below their TTM averages, respectively.

  Risks

  AI capex, the advertising cycle, regulation, and competition may suppress EPS realization.

• Digital Advertising

  A core internet sub-sector, including GOOGL, META, SNAP, PINS, RDDT, and others.

  Strengths

  Large-cap platforms still possess advertising scale and data advantages.

  Weaknesses

  One-week market-cap-weighted performance was about -7.3%, with GOOGL and META acting as major drags.

  Comparison

  Relative to travel at about +4.9% and the shared economy at about +6.0%, digital advertising was weaker in the short term.

  Risks

  Advertising demand volatility, changes in AI search/content distribution, platform competition, and regulatory pressure.

• Travel and Shared Economy

  The relatively strong group in the report, including ABNB, BKNG, EXPE, UBER, DASH, CART, LYFT, and others.

  Strengths

  Travel was about +4.9% and the shared economy about +6.0% on a market-cap-weighted basis over one week, indicating stronger investor preference.

  Weaknesses

  If SBC is treated as a cash cost, EV/EBITDA multiples for the related sectors rise by an average of about 44%, requiring a reassessment of valuation quality.

  Comparison

  Relative to the overall internet sector at -6% and digital advertising at -7.3%, this group significantly outperformed.

  Risks

  Consumer cycle risk, weaker travel demand, platform subsidy competition, SBC, and regulatory risks.

• Gaming/Mobile Applications

  A high-valuation and SBC-sensitive group within internet coverage, including APP, EA, RBLX, TTWO, and others.

  Strengths

  Some companies have content, user scale, and mobile advertising monetization capabilities.

  Weaknesses

  RBLX was about -8% over the week, and SBC cash treatment would raise average EV/EBITDA by about 22%.

  Comparison

  For gaming, investors need to consider revenue growth, EBITDA, and SBC-adjusted earnings quality together rather than relying only on headline multiples.

  Risks

  Slowing user growth, content cycles missing expectations, SBC dilution, and valuation compression.

Key data

• Report Date2026-06-30From the file date and report cover information.
• Coverage RegionNorth AmericaThe report title and industry coverage both point to North American internet.
• Industry ViewAttractiveThe cover shows the Industry View as Attractive.
• One-week Internet Sector Performanceabout -6%SPX/NDX were about -2%/-5% over the same period.
• One-week Performance of Major LeadersGOOGL about -8%, AMZN about -5%, META about -5%The report body notes that the decline was led by GOOGL, with AMZN and META also falling.
• AMZN/GOOGL/META 2026E EPS Multiplesabout 25x/23x/17xRespectively about 22%/11%/20% below their TTM averages.
• Digital Advertising Market-cap-weighted One-week Performanceabout -7.3%The table shows overall pressure on digital advertising companies such as GOOGL, META, SNAP, PINS, and RDDT.
• Travel Market-cap-weighted One-week Performanceabout +4.9%The ABNB, BKNG, and EXPE group outperformed the broader internet sector.
• Shared Economy Market-cap-weighted One-week Performanceabout +6.0%The UBER, DASH, CART, and LYFT group performed relatively strongly.
• NTM EV/EBITDA vs. Historical Averagesabout -10%/-16% vs. 5-year/10-year averagesThe report chart title shows this valuation discount.
• NTM EV/Sales vs. Historical Averagesabout +13%/+16% vs. 5-year/10-year averagesThe report chart title shows sales multiples still above historical averages.
• Impact of SBC Cash Treatment on Gaming EV/EBITDAaverage about +22%The report states that treating SBC as cash would increase EV/EBITDA multiples in gaming.
• Impact of SBC Cash Treatment on Travel/Shared Economy/Real Estate Technologyaverage about +44%The report states that under this measure, EV/EBITDA multiples for the relevant sectors rise even more.

Impact & implications

The investment implication of the report is that the internet sector in the short term is being driven jointly by AI fund flows, valuation repricing, and EPS revisions. If AI investment can generate clearer revenue growth or cost efficiency, leaders such as AMZN, GOOGL, and META may attract capital back through EPS improvement; but if AI capex, competition, and SBC pressure erode profits, the current valuation discount may not translate into upside. At the sub-sector level, the short-term strength in travel and the shared economy shows that capital has not fully exited the internet space, but is instead being reallocated based on earnings visibility and valuation quality.

Risks

• AI-related revenue or efficiency gains may fail to convert into EPS in time, undermining the logic for capital to flow back into the internet sector.
• Although some earnings multiples in the internet sector are below historical averages, EV/Sales remains above 5-year and 10-year averages, so valuation compression risk remains.
• If SBC is treated as a cash cost, valuations in some sub-sectors would rise materially, and earnings quality may be lower than headline metrics suggest.
• Digital advertising, e-commerce, gaming, travel, and the shared economy have different fundamental cycles, which may increase rotation within the sector.
• The report discloses that Morgan Stanley has investment banking, shareholding, or other service relationships with multiple covered companies, and investors should make independent judgments in light of these disclosures.

What to watch

• The direction of EPS revisions for AMZN, GOOGL, and META, and the actual contribution of AI investment to margins.
• Whether internet NTM EV/EBITDA continues to trade at a discount to historical averages, or is passively lifted by earnings downgrades.
• Whether the fact that NTM EV/Sales is above historical averages is absorbed through revenue growth.
• Whether the relative one-week and one-month performance of digital advertising versus travel/shared economy continues.
• Changes in EV/EBITDA under SBC cash treatment, especially for gaming, shared economy, and real estate technology companies.
• Whether stocks with high short interest see fundamental delivery or a reverse squeeze.