E101: Ye acquires Parler, Snap drops 30%, macro outlook, VC metrics, valuing stocks & more

The prediction claimed that in the following years, social‑media 'town square' platforms (Twitter, Facebook, YouTube, etc.) would not remain unchallenged monopolies and that there would be ongoing, viable competitors (including new ones), showing these services are 'just application layers' with alternatives.

Evidence since late 2022 supports this:

• Twitter/X now faces multiple direct text‑based rivals. Meta launched Threads in July 2023 explicitly as a “Twitter competitor,” quickly surpassing 70 million signups in its first days and later reaching hundreds of millions of active users, making it a serious alternative for news- and text‑centric social posting. (cnbc.com) Bluesky, launched in 2023 and opened fully in 2024, similarly positions itself as a micro‑blogging rival to X and has grown to millions of users and significant institutional adoption (e.g., the European Central Bank joining to diversify away from X). (en.wikipedia.org) Alt‑tech platforms like Truth Social also continue to operate with millions of monthly active users as Twitter/X alternatives, particularly for specific political communities. (en.wikipedia.org)

• YouTube and Facebook never achieved uncontested 'town square' monopolies. Video and social posting are now heavily fragmented: YouTube competes with TikTok and Rumble for creator and viewer attention, while Facebook competes with Instagram, TikTok, Snapchat, Reddit, and others. A recent Pew‑style snapshot shows U.S. adults using a range of platforms—YouTube, Facebook, Instagram, TikTok, WhatsApp, Reddit, X, Threads, Bluesky, Truth Social—rather than any single dominant 'town square.' (thedailybeast.com)

• Structural implication – no unchallenged monopoly. By 2025, for the core 'town square' use case (public, text‑ or video‑based discourse at scale), users and institutions can and do choose among several large networks (X, Threads, Facebook, Instagram, TikTok, YouTube) plus niche but persistent alternatives (Mastodon, Bluesky, Truth Social, Rumble, others). This matches Friedberg’s claim that what were once seen as monopolistic, infrastructure‑like 'town squares' are in practice application‑layer services subject to competitive entry and user switching.

Given the presence and growth of multiple credible competitors across the major social‑media categories, and clear erosion of any notion that one firm holds an unchallenged monopoly in the 'town square' space by 2025, Friedberg’s prediction is best judged as essentially correct.

Available evidence three years after the October 22, 2022 episode does not match Chamath’s prediction that Snap would be "abandoned" by investors and become a thinly owned "refugee" in the public markets.

Key points:

• Institutional ownership remains large, not abandoned. Multiple datasets show roughly 50%± of Snap’s shares are still held by institutions, with hundreds of institutional holders:

  • MarketBeat reports institutional ownership of 47.5% as of late November 2025, with substantial inflows and outflows over the last 24 months, not a collapse of interest. (marketbeat.com)
  • Tickergate and other trackers similarly show around 51% institutional, ~24% insiders, ~25% retail. (tickergate.com)
  • Fintel lists ~827–829 institutional owners holding about 865–867 million shares (~59% of shares, ex‑13D/G), i.e., a very broad shareholder base. (fntl.au)
  • BusinessQuant notes institutions owned 51.8% of shares as of June 2025, down only modestly from 53.9% in March, with the institutional owner count down ~6% year‑over‑year—rebalancing, not wholesale flight. (businessquant.com)

• The stock is heavily traded, not “thinly owned” or illiquid.

  • Investing.com shows an average daily volume around 30M shares; YCharts shows a 30‑day average volume of ~129M shares in October 2025. (investing.com)
  • MarketWatch repeatedly reports single‑day volumes well above 100M–250M shares (e.g., 266M shares on Sept. 30, 2025), far exceeding what would be considered thin trading. (marketwatch.com)
  • Ainvest notes daily trading turnover around $280–290M, ranking Snap roughly 350th–380th by U.S. trading volume—again, very ordinary for a mid‑cap tech stock, not a stranded “refugee.” (ainvest.com)

• Share count has grown, not shrunk, and large holders remain.

  • Shares outstanding rose from about 1.61B in 2022 to 1.66B in 2024 and ~1.70B in 2025, meaning the absolute number of shares held by public investors has grown, not contracted. (macrotrends.net)
  • Major asset managers (FMR, Vanguard, Capital World, BlackRock, etc.) continue to hold tens or hundreds of millions of shares each. (mlq.ai)

• Coverage remains broad; the stock is out of favor but not orphaned.

  • Around 30–44 sell‑side analysts still cover Snap, with a consensus rating of Hold and explicit target prices—hardly the profile of a stock that institutions have “abandoned.” (pricetargets.com)
  • News and research from Barron’s, Investors Business Daily and others routinely discuss Snap’s earnings, ad‑market exposure, and AI or product catalysts, indicating ongoing institutional and media attention. (barrons.com)

• Governance concerns remain, but they did not cause mass exit.

  • Snap’s founders Evan Spiegel and Bobby Murphy still control >99% of voting power; Class A shares have no voting rights, exactly the governance structure Chamath objected to. (sec.gov)
  • However, despite this entrenched control, the data above show continuing large‑scale institutional ownership and active trading, contradicting the idea that investors would broadly conclude “there’s no point” and leave the stock stranded.

• Price performance is weak but that alone doesn’t fulfill the prediction.

  • By late 2025 the stock trades in the high single digits, well below both its 2021 peak and its 52‑week high (~$13–17), and has suffered several double‑digit post‑earnings drops. (investing.com)
  • That supports “out-of-favor” sentiment, but not the stronger claim that the shareholder base would shrink to the point of being a thinly owned “refugee.” Ownership and liquidity metrics clearly do not show that outcome.

Putting this together: governance remains founder‑controlled and the stock is volatile and unpopular at times, but institutions still own roughly half the company, trade it heavily, and dozens of analysts cover it. There is no evidence of the kind of sustained, near‑total investor abandonment and thin ownership Chamath predicted. Hence the prediction is best classified as wrong overall.

Meta has, by late 2025, clearly adopted the kind of cost-discipline and shareholder-return playbook Chamath was referring to, and its business and stock have moved back into line with the mega-cap tech cohort.

1. Cost discipline and margin recovery
   After its 2022 slump, Meta pivoted hard to efficiency. In early 2023 Mark Zuckerberg declared 2023 the company’s “Year of Efficiency,” cut more than 20,000 jobs, and lowered its 2023 expense outlook by about $5 billion while also trimming capex guidance. (cnbc.com) By Q3 2023, Meta’s operating margin had doubled to around 40%, with Q4 2023 EBITDA margin near 47% and net margin around 29%, reversing the prior margin compression and putting profitability back in the same high‑20s/low‑30s band typical of the “Magnificent Seven” mega‑caps. (investing.com) This is exactly the sort of cost discipline and margin focus that defined the mature-playbook at Microsoft, Alphabet, and Apple.

2. Shareholder‑friendly capital allocation (buybacks and dividends)
   Meta also embraced the capital‑return behavior of its peers. In February 2023 it authorized a new $40 billion share repurchase program alongside its cost‑cut plan. (euronews.com) In early 2024 it went further, announcing its first ever cash dividend ($0.50 per share quarterly) and authorizing an additional $50 billion of buybacks, explicitly aligning its policy with other “Magnificent Seven” names that routinely return large amounts of cash to shareholders. (business-standard.com) That shift from pure reinvestment to a balance of investment plus dividends/buybacks closely matches the Microsoft/Apple/Alphabet playbook Chamath had in mind.

3. Business and stock performance reverting toward the large‑cap tech mean
   Operationally, Meta’s ad business re‑accelerated, with 2023–2024 revenue and profit growth sharply higher on the back of both ad demand and lower costs, and its Family of Apps segment once again generating very high margins. (cnbc.com) The stock responded: Meta gained roughly 178% in 2023 alone—its best year ever and one of the top returns in the S&P 500—erasing much of the 2022 collapse and returning it to the front rank of mega‑cap tech. (cnbc.com) By 2024–2025, it was firmly a core member of the “Magnificent Seven”/trillion‑dollar club, with overall group net margins in the high‑20s and Meta’s own margins and growth profile in that same neighborhood. (kiplinger.com) One analysis even shows that since 2022, Meta’s total return has been almost identical to an equal‑weighted basket of the seven mega‑caps, i.e., its stock performance has effectively converged to the peer average. (priceactionlab.com) Valuation metrics based on forward cash flow also now place Meta in the same broad range as Alphabet and below (cheaper than) Apple and Microsoft, consistent with it being viewed as part of the same mature, cash‑generative cohort rather than an outlier. (fool.com)

Given this evidence—aggressive cost cuts and efficiency focus, large‑scale and ongoing buybacks plus a new dividend, and business/stock performance that has moved back into alignment with the mega‑cap tech group—the prediction that Meta would "revert to the mean" of peers like Microsoft, Google, and Apple has, by late 2025, effectively come true.

So far, the interest-rate condition of Chamath’s prediction has largely been met, but the governance outcome he forecast is not clearly borne out yet, and the 4–5 year window he specified has not fully elapsed.

1. Rates have been in the 3–5%+ zone for most of the window so far. After his October 2022 comment, the Fed funds rate rose from ~3–4% to 5.25–5.50% in 2023 and then was cut to the 4–4.5% range by late 2024–2025, remaining far above the zero-rate era he was contrasting with and broadly within/around his “3–5%” band.(en.wikipedia.org)

2. Dual‑/multi‑class and supervoting IPOs have continued in this higher‑rate regime.

   • The Council of Institutional Investors notes that dual‑class structures already made up ~24–25% of US IPOs by 2021, and that the proportion of IPOs with differential voting shares has risen since 2019 rather than fallen.(cii.org)
   • A 2024 analysis finds that, by 2024, the share of US IPOs with dual share classes had roughly tripled to about 30% versus a 40‑year average of 10%, indicating the structure is still widely used despite higher rates.(etoro.com)

3. Recent high‑profile IPOs show strong founder/control structures still getting done. Examples since 2023 include:

   • ODDITY Tech (2023): IPO with Class A (1 vote) and Class B (10 votes) shares; post‑IPO, Class B represents ~72% of voting power and the CEO alone holds ~76% of total voting power.(sec.gov)
   • Reddit (2024): Three‑class structure (A: 1 vote, B: 10 votes, C: no votes). After the IPO, Class B holders control ~97% of voting power, and CEO Steve Huffman controls about three‑quarters of total voting power via voting agreements, qualifying Reddit as a NYSE “controlled company.”(sec.gov)
   • Figure Technology Solutions (2025): Dual‑class structure where founder Michael Cagney retains voting control via 10‑vote Class B shares.(reddit.com) These are exactly the sort of aggressive founder‑control/supervoting IPOs Chamath suggested would be checked by investor alternatives in a 4–5% rate world, yet they have been underwritten and completed.

4. There is growing investor pushback, but it hasn’t clearly translated into “meaningful resistance” that stops such IPOs.

   • Large institutions and groups like CII and the Investor Coalition for Equal Votes have intensified campaigns against perpetual dual‑class structures and pushed for time‑based sunsets.(cii.org)
   • However, empirical tracking by CII and others shows the prevalence of dual‑class IPOs has stayed elevated or increased; there is no clear drop in use that could confidently be attributed to higher rates acting as a binding “check.”(cii.org)

5. The forecast horizon is not over, and key IPO candidates haven’t tested the claim yet. Chamath spoke of the “next 4 or 5 years” from late 2022; as of November 2025 we are only about three years in. Several of the most important governance test cases (e.g., Stripe, Databricks and other late‑stage unicorns) have not yet IPO’d, so we don’t know whether they will be forced to moderate control structures at the point of going public.(en.wikipedia.org)

Because:

• the rate condition has mostly held,
• aggressive founder‑control IPOs are still being executed (suggesting his “that dog doesn’t hunt” claim is not clearly correct so far), but
• the 4–5 year evaluation window has not finished and many pivotal IPOs are still ahead,

it is premature to definitively score the prediction as right or wrong. The evidence to date leans against his thesis that higher rates alone would significantly curtail governance overreach in IPOs, but not strongly enough, nor over a long enough period, to move beyond an inconclusive (too early) judgment.

Evidence from 2022–2025 shows that publicly traded small- and mid-cap companies without clear paths to profitability were indeed heavily repriced and left at depressed valuations, consistent with Friedberg’s prediction.

• Unprofitable tech and SaaS were crushed starting in 2022. During the 2022 growth-stock rout, the Goldman Sachs Non‑Profitable Technology Index fell nearly 70% from its peak to the end of August 2022, far worse than the broader market, as investors fled high-growth, money‑losing tech names. (alliancebernstein.com) Cloud/SaaS names saw a similar reset: the BVP Nasdaq Emerging Cloud Index’s EV/sales multiple dropped from ~15.2× in November 2021 to 4.7× by November 2022, a ~70% compression, driven by higher rates and risk-off sentiment. (nasdaq.com) A separate Nasdaq/WisdomTree analysis noted these cloud valuations fell from 12–14× EV/sales in 2020–21 to roughly 4–5× as the Fed tightened policy. (nasdaq.com)

• Across the market, unprofitable firms badly lagged profitable ones. An analysis of all NYSE and Nasdaq companies over the past five years found that unprofitable firms returned a median ~4.2% annualized versus 16% for profitable companies, with much higher volatility and greater sensitivity to Fed rate hikes (loss-makers dropped about 2.2% on average in the 30 days after each hike, vs. 0.8% for profitable peers). (wealthprofessional.ca) This is precisely the pattern of “stocks being punished” for lacking earnings and cash‑generation.

• Small- and mid-cap indices with many unprofitable names remain discounted. Since the end of 2020, the Russell 2000 (U.S. small caps, many unprofitable) has returned only about 4% annualized and trades around 16× forward earnings—roughly a 30% valuation discount to the large‑cap Russell 1000, a gap that has persisted since 2022. Over one‑third of Russell 2000 constituents were still unprofitable as of mid‑2025. (reuters.com) Recent commentary notes that, even after a 2025 rebound led by unprofitable small caps, small-cap valuations adjusted for losses are still near 25‑year lows. (investopedia.com)

• SaaS revenue multiples have structurally reset lower. A 2023 SaaS valuation study shows U.S. SaaS revenue multiples down about 68% from their December 2021 peak (18.34×) to 5.84× by Q3 2023, underscoring how public cloud/SaaS companies—many of them small or mid cap and not yet profitable—have been repriced to far lower levels. (vestbee.com)

Taken together, from 2022 through at least late 2025, unprofitable or cash‑burning small- and mid-cap companies across tech/SaaS and other sectors saw sharp price declines, major multiple compression, and, in aggregate, persistently depressed valuations relative to both their own 2021 levels and to profitable peers. That matches Friedberg’s forecast that those unable to show a credible path to positive earnings and cash generation would “end up in the shitter,” so the prediction is best classified as right.

Available data conflicts with the timing part of Friedberg’s claim but is not sufficient to evaluate the full 10‑year forecast.

What he predicted (Oct 2022)
He said we were “probably at peak capital being managed in venture funds right now” and that this capital would “likely decline for the next decade.” Interpreted as in the normalized version, this means:

1. 2022 was roughly the cyclical peak in venture-capital assets under management (AUM).
2. Over roughly 2022–2032, total VC AUM would generally decline rather than grow.

What has happened so far
Independent data suggest that 2022 has not been the high-water mark (at least so far):

• Preqin reported global venture capital AUM of about $2.45 trillion in March 2022.(technode.global)
• The Preqin 2025 Global Report: Venture Capital shows VC AUM reaching about $3.1 trillion by Q1 2024, a substantial increase from 2022, even though fundraising and exits have slowed.(crowdfundinsider.com)
• A separate 2024 review of the U.S. VC market cites $1.3 trillion in U.S. VC AUM in 2024, indicating continued growth in capital under management despite a sharp drop in new fundraising and more difficult exit conditions.(namaventures.com)

These figures imply that capital managed in venture funds has continued to rise since 2022, so the narrow statement that “we’re probably at peak capital right now” has already been contradicted by later data.

Why the overall verdict is still “inconclusive”
However, Friedberg’s broader forecast concerned a decade‑long cycle (through ~2032) in which VC capital would generally decline rather than grow. While the early years (2022–2024) show that AUM kept increasing and therefore do not support an immediate prolonged decline, they also do not rule out a scenario in which:

• AUM rolls over later in the decade, and
• By 2032, total VC AUM is lower than in 2022, or the overall 2022–2032 path is net‑downward.

Because we are only about three years into a ten‑year horizon, and AUM levels and macro conditions can change materially over the remaining period, we cannot yet say definitively whether the full decade‑long prediction (about the long‑run direction of VC AUM) is right or wrong. The early evidence challenges his claim that 2022 was the peak, but does not conclusively falsify the 2022–2032 directional call.

Therefore, the appropriate classification as of late 2025 is “inconclusive (too early)”, with the caveat that interim data so far run against his assertion that 2022 marked the peak in venture capital AUM.

As of November 30, 2025, the top U.S. federal marginal income tax rate on individuals is 37%, unchanged from the Tax Cuts and Jobs Act structure that took effect in 2018, with an additional 3.8% Net Investment Income Tax applying to certain investment income for high earners (still far below 60% on its own). This is confirmed by current IRS tax rate tables and summaries of federal individual income tax brackets, which show 37% as the highest marginal bracket for 2024–2025.

Friedberg’s prediction was time‑bounded: he said that at some point within roughly the next 15–20 years from 2022 the top marginal federal tax rate on the wealthiest Americans would exceed 60%. The window he described runs approximately from 2022 to sometime between 2037 and 2042. As of late 2025:

• The specified period has not yet elapsed (we are only ~3 years into a 15–20 year horizon).
• The predicted event (a >60% federal top marginal rate) has not happened yet, but it still could happen within the remaining 12–17 years of his time window.

Because the deadline for the prediction is far in the future and the condition could still be met, we cannot say the prediction is right or wrong yet; it is simply too early to judge.

The prediction concerns a 15–20 year horizon from 2022, i.e., roughly 2037–2042. As of the current date (late 2025), we are only about three years into that window, so it is too early to judge whether U.S. federal tax rates on high earners will eventually reach levels that, in 2022, would be viewed as “exorbitant.”

Public data show that the top federal marginal income tax rate in 2022 and 2025 is 37%, similar to other recent years, and there has not yet been a dramatic shift to much higher, historically extreme rates (e.g., the 70–90% brackets seen in parts of the mid‑20th century). However, the prediction is explicitly about what must happen by around 2040 for the U.S. to continue to thrive, not about what would occur by 2025. Since that period has not elapsed, the correctness of the forecast cannot yet be determined.

Therefore, the appropriate classification now is “inconclusive (too early)”, not “right” or “wrong.”

As of 30 November 2025, the U.S. federal debt‑to‑GDP ratio is well below 200%, let alone 300%. Recent estimates put U.S. federal debt held by the public at roughly 97–100% of GDP, and gross federal debt (including intragovernmental holdings) around 120–130% of GDP, depending on the exact measure and source used, but in all cases clearly under 200%.

Since Chamath’s prediction explicitly describes a future in which U.S. debt‑to‑GDP eventually exceeds 200% and then 300% without causing systemic collapse, the key conditions (surpassing those levels) have not yet occurred, and there is no specified deadline by which they must occur. Therefore, there isn’t enough information yet to determine whether the prediction will ultimately be right or wrong.

Because:

• The debt ratio has not exceeded 200% as of late 2025, and
• The prediction is open‑ended in time ("we’ll eventually go past 200… then 300%"),

the correct evaluation today is that it’s too early to tell whether the prediction will come true.

The prediction window is 10–15 years from 2022, i.e. 2032–2037. As of today (2025), we are only about 3 years past the prediction date, well before the earliest point (2032) at which the prediction is supposed to be evaluated. Therefore, it is too early to determine whether “most of the milk you buy at the store will be identical to cow’s milk [via biotech production]” will come true or not.

Because the specified time period has not elapsed, the outcome of the prediction is inconclusive at this time.

As of November 30, 2025, biotechnological methods such as precision fermentation still account for only a very small fraction of global animal-protein production, which is overwhelmingly dominated by conventional livestock (meat, dairy, eggs, and aquaculture).

Industry and market analyses of alternative proteins consistently show that:

• Global meat, dairy, and egg supplies are still produced almost entirely via conventional animal agriculture, with alternative proteins (including plant-based, fermentation-derived, and cultured meat) making up only a low-single-digit percentage of the total protein market by volume and value.
• Precision fermentation today is primarily used for high-value ingredients (e.g., specific dairy proteins like whey or casein, enzymes, specialty fats), not bulk replacement of all major animal-protein categories. Its production volumes are orders of magnitude smaller than those of conventional livestock.

Because the original claim is about the future (that precision fermentation will become the primary means of producing animal proteins and displace livestock as the dominant source), enough time has not passed to determine whether this long-term structural shift will ultimately occur. We can only say it has not happened yet, not that it never will.

Therefore, the prediction cannot currently be judged as right or wrong; it is simply too early to tell, so the appropriate classification is inconclusive.