The Anatomy of the Bear Ch. 1: The Regime You Are Drawing From

Investment Background

Russell Napier did something no other author in this library did. He went to a library.

Between 1995 and 2005 he read the daily Wall Street Journal through four periods: 1921, 1932, 1949 and 1982. Not the summaries written afterwards. The paper as it landed on a desk, on a morning when nobody yet knew what the following decade would hold. Roughly seventy thousand articles.

The question he was answering is one this library has never asked. Every book here can explain a market bottom. None of them can tell you what a bottom looked like to somebody standing in it — and those are different objects. A chart with the low circled is an artefact of hindsight. The morning newspaper is what was actually available.

This chapter establishes the time scale that makes the rest of the book necessary, because it is a time scale the library currently has no shelf for.

The Wall Street Translation

Three Clocks, and the Missing One

The library measures markets on two clocks, and the gap between them is where a retirement actually happens.

The missing row is the one that matters most to a person with a finite life. Ten to twenty years is long enough to contain an entire accumulation phase, or an entire retirement, and short enough that it is emphatically not "the long run" that any average describes.

Marks's thermometer will move several times inside one Napier regime. That is not a criticism of Marks — the two instruments answer different questions and both readings can be correct simultaneously. You can be at a cyclical high inside a secular bear, which is precisely what 1930, 1937, 1968 and 1972 were. A reader who owns only the cyclical instrument will read those moments as tops to be sold, and will be right; a reader who owns only the regime instrument will read them as noise inside a longer condition, and will also be right. Confusing the two is the error this chapter exists to prevent.

What a Regime Actually Is

A secular regime is not a trend and it is not a forecast. It is a persistent condition governing what a given valuation reading means.

The clearest way to see it is in what the market did not do. Between 1966 and 1982, the Dow Jones Industrial Average went essentially nowhere in nominal terms across sixteen years. In real terms it lost roughly seventy percent of its purchasing power. There were tradable rallies inside that span — several of them large. There was no escape from the condition.

The same shape appears at the other end. From 1982 to 2000, a portfolio that did nothing at all compounded at a rate that made almost every tactical decision irrelevant. The dominant variable in both cases was not skill, not selection, not timing. It was which regime the investor happened to be standing in.

This is the uncomfortable observation at the centre of the book: for a person with one retirement, the regime they draw from may matter more than every decision they make inside it.

The Four Bottoms

Napier's dataset is four dates, and the smallness of that number is the honest starting point for everything that follows.

Four observations is not a statistical sample and this book will never pretend otherwise. What four observations can support is a description of what these episodes had in common — and ch06 states plainly which parts of that description are robust and which are storytelling.

Note what every one of these dates is: the beginning of an enormous bull market. This is not a book about crashes. It is a book about the conditions under which the best available prices in a generation appear, and about why almost nobody recognised them at the time.

Division of Labor With the Rest of the Library

The boundary with boom-and-bust is worth stating early because it is counterintuitive. That book is about the top. This book is about the bottom, and the two are not opposite ends of the same event. A bust is an episode — it happens, it is violent, it is over. A bottom is a condition arrived at years later, after the bust has been forgotten and replaced by exhaustion. In 1932 the crash was three years in the past. In 1982 the trouble had been going on for sixteen years. Nobody at a real bottom experiences it as drama.

Executable Trading Rules

1. Write down which regime your plan's return assumption came from. If your spreadsheet uses a real equity return of six or seven percent, find out what span of history that number was fitted to. Almost every planning figure in common use is drawn from a period containing at least one great secular bull market. That is a choice about sampling, and most people have never made it consciously.

2. Hold the cyclical and secular instruments separately, and never let one overrule the other. Marks's temperature reading tells you about the next three years. A regime reading tells you about the next fifteen. Both can say different things at the same moment without either being wrong, and a plan needs both answers for different purposes.

3. Stop treating a flat decade as a malfunction. Sixteen years of nominal stagnation is inside the historical record, not outside it. A plan that implicitly assumes such a period cannot occur is a plan that has excluded a documented outcome by assumption rather than by evidence.

4. Do not convert any of this into a timing programme. Nothing in this chapter says a regime can be called in advance, and chapter 6 lays out the evidence that it usually cannot. The purpose of regime awareness is to widen the range of outcomes your plan is built to survive, not to narrow it to a bet.

5. Keep the core untouched while you think about this. A low-cost, globally diversified index core is the right answer in every regime described in this book, including the bad ones. What changes across regimes is what you may safely withdraw from it and how large a buffer sits in front of it, which chapter 5 develops in full.

Relevance to a Retirement Portfolio

The four percent rule is the clearest illustration of this chapter's argument, and retirement-decumulation-mechanics chapter 1 already says the important part: it is an empirical statistic, not a law.

A statistic derived from a sample carries the sample's regimes inside it. The historical runs that produced the four percent figure include the 1949 to 1966 expansion and the 1982 to 2000 expansion — two of the most favourable valuation regimes ever recorded. They also include the 1966 starting cohorts, which is why the number is four and not six. The rule is not naive. It is simply a summary of a particular run of monetary and valuation history, and a secular regime is precisely the kind of object that can put you outside it.

This does not mean the rule is wrong or should be discarded. It means the honest way to hold it is as a central estimate with a regime-dependent spread, and that the correct response to that spread is the machinery retirement-decumulation-mechanics already provides: guardrails in chapter 4, a buffer in chapter 3, sequencing in chapter 5. This book does not propose a replacement for any of it. It argues that the size of the uncertainty those tools are absorbing is larger than a single number suggests, and that the reason is structural rather than statistical noise.

And the standard recommendation is unchanged: a core of low-cost, globally diversified index funds, no leverage, a cash buffer covering essential spending, and withdrawal rules containing an adjustment mechanism. Nothing in this book displaces indexing. A regime is a statement about the distribution you are drawing from, not an instruction to trade — and a reader who leaves this chapter planning to move to cash has read the opposite of what it says.

Chapter 2 turns to what the four bottoms actually looked like from inside, on the mornings nobody knew what they were standing in.