> For the complete documentation index, see [llms.txt](https://laurence-wilse-samson.gitbook.io/textbooks/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://laurence-wilse-samson.gitbook.io/textbooks/financial-economics-claims-prices-holders/readme.md). # README Laurence Wilse-Samson — NYU Wagner School of Public Policy () A one-volume course text in financial economics for advanced undergraduates and for masters students in finance, economics, and public policy, assuming one statistics course at the level of multiple regression and calculus through partial derivatives and nothing else. Six parts, twenty-seven chapters, four appendices. It is one volume rather than two because the questions do not divide along the usual seam: asset pricing and corporate finance are taught in different rooms, and the effect is to hide that the same claims appear in both — a firm designs and issues a security, an investor with a mandate and a capital charge buys it, and the price is a joint product of what the security promises and of what its buyer is permitted to hold. The subtitle names the spine the book follows: the claims, their prices, and the holders who end up owning them. Its organizing question is whether the third of these belongs in the theory of the first two, and its answer is that it does, that this is now a measurable proposition rather than a rhetorical one, and that the measurement is the interesting part. PhD-preparation material is carried in starred sections the body never depends on; Chapter 1 §1.5 states that convention and `frontmatter/FOR_INSTRUCTORS.md` prints two thirteen-week syllabus maps. | Part | Chapters | | ----------------------------- | -------- | | I. Foundations | 1-3 | | II. Asset Pricing | 4-8 | | III. Asset Markets | 9-13 | | IV. The Investor Ecology | 14-20 | | V. Firms as Issuers of Claims | 21-25 | | VI. Risk and the Real Economy | 26-27 | ## Repository map | Path | What it holds | | -------------------------------- | ----------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | | `chapters/` | The 27 chapters, one file each, `chapter_NN_slug.md`, one H1 per file | | `frontmatter/PREFACE.md` | The preface | | `appendices/` | Appendices A-D: `appendix_A_empirical_methods.md`, `appendix_B_data_sources.md`, `appendix_C_accounting.md`, `appendix_D_math_review.md` — text complete; their figures are the open item | | `backmatter/` | Consolidated glossary, bibliography and index, all three **generated** from the chapters by `code/build_backmatter.py` and never edited by hand | | `literature_map/` | Five memos mapping the author's literature collection to the drafted chapters | | `SUMMARY.md` | GitBook table of contents — the book in reading order | | `OUTLINE_v3.md` | Outline of record: parts, chapter briefs, word budgets, chapter template, boundary rules | | `NOTATION.md` | Symbol registry — check any symbol against it before introducing or reusing one | | `REVIEW_FLAGS.md` | Open author items: data to verify, word budgets, sourcing to confirm, structural calls | | `REVIEW_2026-09-01.md` | Scorecard and work plan: where the manuscript stands against a 9.5, and the ordered list of what closes the gap | | `FIGURES_AND_APPENDICES_PLAN.md` | Exhibit plan (221 numbered exhibits, 137 of them P1) and the section outlines for Appendices A-D | | `code/` | Figure and table scripts: the `fe_figures/` package plus per-chapter `chNN_figures.py` — pipeline being stood up | | `figures/` | Rendered exhibits: `png/` for GitBook, `pdf/` for typesetting | Where the outline and a chapter file disagree on scope, `OUTLINE_v3.md` wins; flag the conflict in `REVIEW_FLAGS.md` rather than reconciling it silently. Local and untracked, deliberately (all of it `.gitignore`d): `literature/` (\~6,650 PDFs, \~3.6GB on Google Drive — never commit it, and never run recursive operations over it), `CLAUDE.md`, `next_session_plan.md`, `BOOK_OUTLINE_PROPOSAL.md` (an obsolete 30-chapter structure — do not write to it), and `chapters/chapter_15_institutional_finance.md`, a superseded draft the current Chapter 15 replaced: it sits in the working directory, but it is not in the repository and is not part of the book. Tracked but superseded, kept only for history: `PORTFOLIO_AND_OUTLINE_v2.md` and `OUTLINE_REVIEW_v2.md`. ## Build GitBook consumes `SUMMARY.md`. Parts are `##` headers; chapters and appendices are links to files. A file not listed there does not appear in the book, which is why the planning docs stay out of it. Add a file to `SUMMARY.md` in the same edit that creates it. Figures render to `figures/png/` (8x5 in, 300 dpi — oversized on purpose, so GitBook downscales cleanly and print needs no re-run), with vector copies in `figures/pdf/`. Reference them with markdown image syntax, never raw HTML, and carry the caption and `*Source: …*` line as text under the image so both stay searchable and survive typesetting. ## Conventions for contributors **Wrapping.** Chapters and frontmatter are soft-wrapped: one line per paragraph, no hard breaks. This is deliberate — every edit then shows as a whole-paragraph diff, which is what prose review wants. Root planning docs, `SUMMARY.md`, and this file hard-wrap at \~78 characters; table rows and link lines run long because they cannot be broken. **Math.** Display math goes in `$$…$$` with the delimiters on their own lines. Never escape underscores inside math — write `r_{t+1}`, not `r\_{t+1}`, since GitBook's KaTeX needs raw underscores for subscripts. Write superscript stars as `^{\ast}`, never `^*`. **Backslash-punctuation inside math is forbidden** — this is load-bearing, not style. Both renderers this book meets (GitHub's file view, MathJax- based, where the manuscript is read today; GitBook/KaTeX, where it may publish) run markdown escape-processing inside math — even own-line `$$` blocks — so a backslash before punctuation is silently eaten: `\left\{` becomes the error `\left{`, `\;=\;` renders as `; = ;`, `\{0\}` silently renders as `0`. Letter-form commands survive, and the dialect below is valid in BOTH MathJax and KaTeX: `\lbrace`/`\rbrace` (never `\{`/`\}`); no `\, \; \! \:` (use `\thinspace` where a gap is load-bearing, `\quad` to separate constraints); `\Vert` (never `\|`); `\cr` (never `\\`) as the row separator in `cases`/`aligned`/`pmatrix`. Do not "fix" these back to the standard LaTeX forms: that re-breaks the render. **No `$` or `%` characters inside math, ever.** There is no portable letters-only glyph for either (`\char` fails on MathJax, `\unicode` fails on KaTeX, `\textdollar` on both, and a bare `%` is a comment character that swallows the rest of the expression). Equations carry bare numbers; units live in prose ("dollars", "percent"); money amounts in prose as `\$100`. Reintroducing the glyphs is a typesetting-time script, not a manuscript convention. **Money.** Escape dollar signs in prose and in table cells: `\$1.2 trillion`. An unescaped `$` opens a math span and swallows the rest of the paragraph, or the rest of the table row. One more trap: a prose `\$` becomes a bare `$` after markdown escaping and can steal the closing delimiter of a LATER inline `$…$` span in the same paragraph — so never place `\$` money earlier in a paragraph than inline math. Where that cannot be arranged, spell the unit: `108 dollars`. **Spaces in front of closing subscripts are load-bearing.** The same markdown-first pass reads underscores as emphasis delimiters. In `$\mathrm{RE}_t = \mathrm{RE}_{t-1}$` the first underscore opens a run and the second closes it, so the parser wraps the middle in ``, eats both underscores, and the paragraph renders as raw TeX with its subscripts missing. A space in front of the *closing* underscore — `\mathrm{RE} _{t-1}` — leaves it unable to close, and TeX ignores the space, so the rendered output is unchanged. The same pass reads `\right](0.015)` as a link, which a space between `]` and `(` prevents. Both spaces look like typos and are not; do not close them up. **No math inside italics.** An inline `$…$` span nested in a `*…*` emphasis span does not render — the line shows as literal source. In source notes and Readings annotations, spell parameters in prose ("with risk aversion 4"), or close the italics before a formula that must stay symbolic. Italic lines carrying only `\$` money are fine. The check\_math lint under `code/` enforces all of these rules. **Punctuation.** Spaced em dashes, `—`. Straight quotes, not curly. **Boxes.** `> **Box N.M — Title**`, then a `>` line, with the whole box blockquoted. **Tables.** A numbered table carries a `**Table N.M: Title**` caption line above it and an italic `*Source: …*` line below it. Every table states its source; every substantive number is traceable. **Notation and flags.** Check every symbol against `NOTATION.md` — drift is the known failure mode in a book this equation-heavy. Anything only the author can settle goes into `REVIEW_FLAGS.md` rather than being guessed; an item is cleared by deleting its line. **Files.** LF line endings, no trailing whitespace, UTF-8. The repository stores LF throughout; a Windows working tree shows the chapter files as CRLF because `core.autocrlf` converts on checkout, which is expected and is not something to correct. Whatever a file already has, keep it: rewriting line endings rewrites the whole file's diff and buries the edit that mattered.