The Road to Serfdom, Part 3: The End of Truth and My Investing Bias

The central answer: In Chapter Eleven of The Road to Serfdom, “The End of Truth,” Friedrich Hayek’s concern is not simply that propaganda exists. It is that when a comprehensive political and economic plan demands agreement on one ultimate system of ends, it may also demand agreement on supposedly factual explanations that legitimate every concrete choice. Evidence then stops being a way to correct the plan and becomes a way to defend a decision already made. This describes an argument made in the context of the 1944 book, not a verdict about any contemporary society or institution. [1][2] My own reported investing return has gone from roughly +40% at its peak to about -5%, yet I resist selling. That is the personal self-confirmation problem this essay examines—not investment advice attributed to Hayek.

Why this follows the chapters on planning and security

This is the third substantive entry in one reading sequence. Part One examined comprehensive economic planning, political authority and the rule of law. Part Two considered Chapter Nine, “Security and Freedom”: a general subsistence floor is not identical to a promise to preserve particular jobs and earnings. Chapter Ten, “Why the Worst Get on Top,” examines the selection pressures that may arise within highly concentrated power structures. Chapter Eleven changes the question: why would authorities also seek control over how people explain the world?

There is a distinction worth preserving. The power to allocate a scarce resource is visible; the power to determine what counts as a success, whose failure data can be cited and which explanations are heard is less visible. These powers can reinforce each other. They are not automatically identical, and the existence of official messaging does not itself establish the presence of comprehensive economic planning.

For this reading, the primary textual reference is the English Chapter Eleven in the 2001 Routledge Classics edition, checked against its reproduced text and the chapter sequence of Bruce Caldwell’s University of Chicago Press edition. The chapter starts at printed page 157 in that Routledge Classics edition; page numbering varies across editions. The 1997 Chinese edition’s bibliographic record confirms the corresponding translated chapter heading. These checks do not establish that XBSTACK’s author owned, annotated or finished reading a particular Chinese printed copy. [1][2][4]

Three short passages worth keeping from the original chapter

A reading note should retain a small amount of the author’s actual language instead of presenting only my interpretations. These are brief, verbatim excerpts from Chapter Eleven, with printed page numbers from the 2001 Routledge Classics edition. They are fragments, not quotations of complete paragraphs, and other editions may paginate them differently. [2]

First: the danger of centrally aligned messaging. Hayek writes that “all propaganda serves the same goal” (p. 158). In context, he is not claiming that persuasion itself is uniquely totalitarian. He is describing what changes when ostensibly different outlets are systematically coordinated to reinforce the same ends. The practical question is whether independently generated disagreement can still be heard.

Second: the manipulation of political vocabulary. He describes the technique to “use the old words but change their meaning” (p. 161). The discussion that follows takes liberty as a central example: a familiar and attractive word can be used for an institutionally different arrangement. The analytical lesson is to define a claim before embracing its emotional associations, not to accuse every present-day institution of the same political behavior.

Third: intellectual growth through disagreement. In defending the role of independent inquiry, Hayek writes: “The growth of reason is a social process” (p. 169). The surrounding argument requires diversity of knowledge and an ability to challenge accepted claims. My personal portfolio experience provides a limited analogy: evidence against my holdings deserves consideration even when I dislike its implications. Hayek did not write an investment stop-loss formula.

An attribution detail: the epigraph at the start of Chapter Eleven is explicitly attributed to E. H. Carr. It discusses the relationship between nationalizing thought and nationalizing industry. It is a quotation selected by Hayek, not a line authored by Hayek. [2]

Persuasion is not the same as a monopoly over persuasion

Hayek begins by describing a dilemma faced by authorities trying to align a whole society with one hierarchy of ends. Coercion may make people comply, but a stable regime would prefer people to believe the ends are their own. This brings him to propaganda. The key distinction in the next pages is easy to miss: persuasive techniques are not, in Hayek’s account, unique to totalitarian government. Schools, advertisers, parties, campaigners and media organizations try to influence views in many political systems. What changes the character of propaganda is the concentration of competing sources under unified control. [2]

That distinction prevents a lazy reading. A public-health campaign can simplify evidence to encourage action. An advertiser highlights the advantages of a product. News organizations employ editorial judgment. Such practices can deserve criticism or stronger scrutiny. But the presence of persuasion is not sufficient to prove the kind of system Hayek described.

His stronger concern applies where public claims become difficult to challenge because alternative sources are excluded, inaccessible or punished. It matters whether independently generated evidence can be published; whether disagreements can survive; and whether the public can evaluate claims that do not come through a single institutional gatekeeper. That is a question about the structure of inquiry, not just the tone of an individual message.

When value judgments turn into compulsory factual claims

An organization may openly say that it prefers one outcome to another: fairer transport access, faster economic growth, financial stability or lower emissions. These are policy or value judgments, although they involve empirical expectations. A different kind of claim asserts that a particular policy will produce those outcomes. That assertion is partly empirical. It can be checked against evidence, competing causal explanations and observed results.

Imagine a city choosing to expand public transport. Advocates may reasonably prefer lower congestion and better access for residents without private cars. Now imagine a stronger claim: that every line will necessarily reduce traffic, that any missed ridership target is impossible or that dissent about costs proves hostility to social equality. The decision’s intended goals are being used as substitutes for the evidence needed to evaluate its actual performance.

This is an invented analytical example, not an episode quoted from Hayek or a description of a real city. It illustrates a mechanism emphasized in Chapter Eleven: the governing system must justify not only the ends it pursues, but the factual and causal propositions on which its specific means rest. When decisions are made before their rationale has been established, a persuasive explanation can be built afterward and defended as part of the official creed. Hayek discusses the use of myths and seemingly scientific theories for that purpose. [2]

The process does not require every participant to consciously lie. A planner may sincerely mistake a preferred explanation for a proven one. An administrator may be rewarded for reproducing success metrics rather than probing their definition. A reporter may lack access to alternative data. These are different motives, but they can reinforce an institutional outcome in which it becomes expensive to challenge the conclusion.

There is an important limit to this reasoning. Disagreeing with a policy does not prove its evidence was fabricated. The relevant questions are whether claims are falsifiable, whether the underlying data are available, whether competing explanations are permitted and what happens when the anticipated results fail to appear.

Seven original terms to understand before drawing modern comparisons

These expressions occur in Chapter Eleven. The explanations retain their historical and argumentative context rather than recasting the book as a work of investment psychology. Printed page numbers refer to the 2001 Routledge Classics edition. [2]

Original term	Meaning and limits in Chapter Eleven
System of ends	From p. 157: a hierarchy of ultimate purposes demanded by comprehensive social planning, not an individual’s ordinary personal goals.
Propaganda	pp. 158–159: techniques of persuasion are not unique to totalitarianism; the concern is the concentration and coordination of information sources.
Gleichschaltung	p. 158: a German political term used here for coordinated alignment or enforced uniformity of minds and propaganda; it does not mean individual investor confirmation bias.
Questions of fact	p. 159: factual and causal claims about possibilities and methods; a preferred end cannot stand in for their verification.
Myth / pseudo-scientific theory	pp. 160–161: rationalizations constructed to make pre-existing ideological preferences or political decisions appear objectively necessary.
Perversion of language	pp. 161–163: familiar words, notably liberty, are redefined while retaining positive associations.
Intellectual freedom	pp. 168–169: freedom for some people to challenge accepted claims publicly and test competing reasons, not the expectation that everyone independently studies every issue.

Modern terms are separate: In describing my year-to-date returns going from roughly +40% to -5% and my reluctance to sell, confirmation bias and a possible disposition effect come from modern behavioral finance. They are not Hayek’s terms in this chapter, and my individual experience does not prove a formal psychological diagnosis. It is a limited analogy about resistance to contrary evidence, not an equivalence between investing decisions and coercive political power.

From a 40% gain to a 5% loss: the facts I would rather explain away

These questions are no longer abstract to me. Looking at the recent fluctuations in China’s A-share market and US stocks, I also have to look at my own portfolio. At its best this year, my year-to-date return was about +40%. As I revise this draft on October 25, 2026, it has slipped to roughly -5% under my own account’s reporting convention. That is a difference of 45 percentage points in reported return, not a verified portfolio maximum drawdown or the price decline of a particular holding. Without independently checked cash flows, transaction records and valuation timestamps, a precise performance attribution would be misleading.

The markets themselves provide a useful reminder of how selective narratives can be. Between October 19 and 23, 2026, the Shanghai Composite lost about 0.69% for the week and China’s ChiNext index lost about 3.18%. Yet on Friday, October 23, the ChiNext index rose 4.16% and the Shanghai Composite gained 0.89%. On that same Friday, the US S&P 500 declined about 0.52% and the Nasdaq Composite about 0.21%. These are separate markets and time horizons, not one story about a synchronized collapse. They also do not explain the change in my portfolio. My individual holdings, position concentration, trading decisions and currency exposures have not been reconciled here. [6][7][8]

The more uncomfortable fact is that after giving back so much of the gain and moving into a loss, I still do not want to sell. When I was up around 40%, it was easy to treat the profit as proof of good judgment: perhaps I understood the companies, the technology cycle and the long-term opportunity better than others. As those gains disappeared, however, keeping the same positions demanded a different kind of evidence. Instead, I find myself tempted by comforting questions: Might the next earnings report be better? Isn’t the industry still growing? Won’t the market bounce back? These are perfectly legitimate questions if they are investigated on their merits. If I only accept answers that support holding, I am no longer testing the investment thesis. I am protecting the feeling that I used to be right.

I used to call this conviction. Now I have to face an uglier possibility: in my case, this is not necessarily discipline or perseverance. It is stubbornness, and an unwillingness to confront failure. Selling feels like signing my name to a confession: the gains I had earlier in the year are gone, my confidence may have rested on weak assumptions, and I am not the investor who can always see the direction clearly. I can endure another red number on the screen more easily than I can say, “I may have been wrong.”

There is a strictly limited analogy here to Chapter Eleven. Hayek’s subject was the use of centralized institutional power to align public beliefs and factual claims with a predetermined political purpose. My problem is an individual’s confirmation bias under market uncertainty. The two have different structures and consequences; the book is not a treatise on stop-loss psychology. What they share as a question for investigation is the temptation to choose a conclusion first and let the evidence serve it afterward. In my case, the conclusion is that I should continue holding. Evidence against that position can become “temporary noise,” while supportive scraps are promoted into proof.

Yet recognizing stubbornness should not turn into another emotional shortcut. A move from +40% to -5% does not, by itself, prove that selling today is optimal. A drawdown could reflect valuation compression, changed fundamentals, market sentiment, or a mixture of these; continued ownership could still be justified by current evidence in some cases. The failure I want to correct is not necessarily the mere fact of continuing to hold. It is treating “I cannot bear to sell” as a sufficient investment reason, even when the rationale may have changed.

Before making the next important position decision, I want to write down what justified the original purchase, which facts remain true, which contrary evidence I have dismissed too quickly, and what would make me change or reduce the position. A particularly difficult test is this: if I held the same amount in cash today, would I buy this security again at today’s price, risk level and portfolio weight? The answers should be connected to business performance, valuation, cash flow, uncertainty and downside tolerance—not merely to the memory of a 40% gain. This is a plan for a better review process, not a strategy that I have implemented, backtested or proved effective.

The distinction I want to keep is simple: “I believed it before” is a fact about the past. “It remains worth believing now” is a claim that needs fresh evidence. What I resist losing may be more than money. It may be the version of myself who once appeared to have made the right call. If protecting that image becomes more important than looking at reality, the account loss will not be the only cost; I will also forfeit the chance to correct my judgment.

Old words, new meanings

Hayek turns to the transformation of familiar political vocabulary. Terms such as liberty, equality and justice can retain their positive associations while being redefined to refer to arrangements with very different consequences. When key terms change meaning without disclosure, two sides can appear to agree while talking about different institutions. [2]

The lesson has a modest application beyond the book’s historical subject. In a software business, “growth” might mean new accounts, weekly retained users, qualified visits, or paying customers. A dashboard may show impressive growth while product completion and repeat use remain weak. Neither set of numbers must be fraudulent: they simply answer different questions. But describing all of them with the same unqualified word can give a misleading picture.

For XBSTACK, this is a relevant discipline of interpretation. “Article published,” “indexed,” “shown in search,” “clicked,” “read by a qualified visitor,” and “helped someone complete a tool task” are distinct observable outcomes. Likewise, a company’s accounting profit is not interchangeable with recurring operating earnings or operating cash flow. The comparison is an editorial extension to modern products and financial research, not a claim that Hayek anticipated web analytics or AI applications.

If a claim matters, define the term, identify its measurement method and ask which plausible alternative definition would change the conclusion. Agreement on a word is no substitute for agreement on what can be observed.

The objection: science and technical achievement did not simply vanish

Chapter Eleven reaches beyond politics into academic and scientific inquiry. Hayek argues that a knowledge system oriented exclusively toward a prearranged social purpose can restrict questions whose answers might undermine the program. His examples include ideologically labeled scientific doctrines in twentieth-century totalitarian settings. The concern is not merely censorship in a narrow sense, but the loss of room for unexpected results and independent criticism. [2]

That argument deserves engagement, not unqualified endorsement. In Liberalism’s Last Man: Hayek in the Age of Political Capitalism (2023), Vikash Yadav criticizes Hayek’s conception of propaganda as too narrow and simplified. His publicly available chapter abstract explicitly notes that totalitarian regimes continued to support basic science and sophisticated technological research. It also considers the continued use of information operations in different forms of political capitalism. [3]

This distinction matters. A regime may achieve significant technical advances while constraining academic independence or public debate in other domains. “Scientific work is politically constrained” does not entail that “all scientific and technological progress has ceased.” A serious study would specify which research questions are restricted, who selects the agenda, how researchers can dissent and what kind of technical performance is still possible.

There are additional risks outside systems of comprehensive state planning. Concentrated private media ownership, distorted platform incentives, commercial public relations, and the funding structure of academic research can also impair the quality of public information. These are not interchangeable with the coercion involved in the book’s account of totalitarian planning. Yet treating privately produced information as automatically reliable would be as careless as treating every public institution as necessarily hostile to inquiry.

The strongest defensible lesson is therefore narrower than a grand ideological conclusion: ask whether an institution permits evidence to contradict the outcome it wishes to demonstrate. That question can be applied across many different institutional designs, with each case needing separate empirical investigation.

A five-question evidence audit inspired by the chapter

Instead of treating “The End of Truth” as a slogan, an independent reader can use it to improve how claims are assessed. The following framework is XBSTACK’s contemporary analytical extension. These are not five rules originally set out by Hayek.

Dimension	Practical question
Purpose	Is this sentence saying what ought to be valued, or what has been shown to happen?
Definitions	Which terms and metrics are being used? Would the claim change under another legitimate definition?
Sources	Do several accounts come from truly independent evidence, or do they all repeat one original assertion?
Counterevidence	What observable result would force the author or institution to revise its conclusion? Can unfavorable data be published?
Correction	Is there a real process for changing course, acknowledging error and retaining a correction history?

Consider a new AI inference framework claiming superior efficiency. A responsible engineering decision requires comparable models, devices, quantization settings, tasks and prompts, and must disclose failures. Performance observed in a vendor demonstration is not a verified improvement on the actual target phone. Likewise, the fact that a financial report displays positive net income does not establish that its principal business is profitable: investment income, valuation changes and nonrecurring items need to be identified.

These examples concern the discipline of evidence, not Hayek’s original policy positions. There is no guarantee that a well-structured evidence audit will produce an objectively perfect answer. Its purpose is to make an attractive but false conclusion easier to detect and revise.

The most useful habit is simple: write down what information would change your mind before searching for information that agrees with you. It creates a practical boundary between investigation and self-confirmation.

Where the one-book sequence goes next

The two questions raised by Chapters Ten and Eleven are connected but different. Chapter Ten considers which behaviors may be rewarded inside highly concentrated power structures; Chapter Eleven considers how such structures can come to treat public knowledge as an instrument of shared purpose. Explaining one does not prove every element of the other.

The reading sequence now has three distinct questions: authority and rule of law, security and freedom, truth and correction. The aim is neither to use a 1944 book as an automatic answer to present-day politics nor to translate its language into contemporary stock recommendations. Claims about any actual institution, company or policy must be checked against its own evidence and alternatives.

Return to this book sequence: Part One: Planning and the Rule of Law → Part Two: Security and Freedom → this essay on truth and correction. The Reading collection lists other books and essays.

Continue with the reading-as-decision framework, the Horizon collection, and the investing research archive. To see what kind of real work anchors the site’s broader interests, visit About XBSTACK.

Sources and verification boundaries

1. Routledge, The Road to Serfdom: Text and Documents—The Definitive Edition, ed. Bruce Caldwell: publisher and verified sequence of chapters. https://www.routledge.com/The-Road-to-Serfdom/Hayek-Caldwell/p/book/9780415755320
2. F. A. Hayek, The Road to Serfdom, Routledge Classics edition (2001), Chapter Eleven, “The End of Truth,” beginning at printed page 157: the chapter was consulted directly in the reproduced English text; no extended copyrighted passages are reproduced here. https://xenopraxis.net/readings/hayek_roadtoserfdom.pdf
3. Vikash Yadav, Liberalism’s Last Man: Hayek in the Age of Political Capitalism (2023), “The End of Truth,” University of Chicago Press / Oxford Academic. The public chapter abstract was verified; the paywalled full chapter was not read. https://academic.oup.com/chicago-scholarship-online/book/55542/chapter-abstract/432547650
4. China Social Sciences Press, bibliographic and contents record of the 1997 Chinese edition of The Road to Serfdom: confirms the chapter name and edition, not a full line-by-line collation of the Chinese translation. https://www.csspw.com.cn/booksdetail_15923_2066240_1.jhtml
5. University of Chicago Press definitive edition, edited by Bruce Caldwell: independent publisher reference. https://press.uchicago.edu/ucp/books/book/chicago/R/bo4138549.html
6. Sina Finance, A-share weekly review for October 19–23, 2026: Shanghai Composite and ChiNext week-on-week moves. https://finance.sina.com.cn/tech/roll/2026-10-23/doc-iniwffek8229274.shtml
7. 21st Century Business Herald via Eastmoney, A-share session close on October 23, 2026: Shanghai Composite and ChiNext one-day gains. https://finance.eastmoney.com/a/202610233906227841.html
8. Reuters, Wall Street closing recap on October 23, 2026: S&P 500 and Nasdaq losses. https://www.reuters.com/business/wall-street-futures-dip-after-ai-rout-telecoms-slide-2026-10-23/

Editorial integrity note: The source-based reading analysis does not claim an annotated or completed reading of a particular printed Chinese edition. The author’s roughly +40% to -5% year-to-date reported return and reluctance to sell are the author’s own newly supplied account, not independently reconciled brokerage transactions or a verified maximum drawdown. Investment-psychology and AI-product parallels are explicitly editorial, not claims that Hayek made them in 1944, and not trade recommendations.