The Financial ReviewDetailed Format and Style Guidelines
Title page and abstractThe first page should contain:
The paper title should be as short and clearly informative as reasonably possible and must not begin with "on" or any other preposition. The second page should repeat the title and also include an abstract of exactly 100 words or fewer. The abstract must not include citations but should focus on what is new in the current paper. If it is important to point out the current paper's extension of or difference from one particular paper, the abstract should refer generically to previous work, a widely cited study, existing evidence, etc. In that case, of course, the introduction definitely should cite the particular paper in question. In writing the body of the paper, do not assume that the reader has read the abstract. After conditional acceptance, move the abstract to the title page and eliminate the separate abstract page. Section headings:Number all major body sections including the introduction. Section headers should be left justified with Arabic numerals. Only the first word of the section title is capitalized. Headers are in boldface and subheads are italicized. For example:
Footnotes:
References:
For periodicals (one author):
For periodicals (two or more authors):
Journal titles are never abbreviated in the reference list. For working papers:
For books:
For internet sites that do not have print equivalents:
French, Kenneth R., 2005. Data Library. For court cases:
Tables, figures, and graphs:Authors should keep in mind that the size of The Financial Review text block is 7-1/8" long and 4-5/8" wide.
All graphs and diagrams should be referred to as figures and should be numbered consecutively in the text in Arabic numerals. The figure number should be left justified and appear under the figure. The figure title is bolded and left-justified and appears directly beneath the figure. Under the title there follows brief descriptive material to allow a reader to understand the figure without reading the body of the paper. Tables should be numbered consecutively in the text in Arabic numerals. The table number is left justified and appears above the table body. Under the table number and also left justified should appear the table title in bold. The body of the paper must discuss each table and panel. Each row and column of each table must receive some discussion, even if only in a summary that encompasses more than one row or column. The body of the paper must not explain in detail how the table is laid out or what its symbols mean; as explained below, the table itself must be designed for maximum clarity without reference to the body of the paper. Please consider splitting long tables into separately numbered tables, instead of making continuation pages. The purpose of a table is to present information in a form that the reader can quickly grasp. Authors should look at all details of table design from the reader's perspective. Table titles should be reasonably short, but long enough to state what the table is about. Avoid the word "results" as it is obvious to any likely reader of the journal that tables present results. Please use relatively specific terms like "regressions." Beneath the table title, there often should follow an explanatory paragraph to allow the reader to understand the table without reference to the body of the paper or another table. (However, it can be appropriate in some papers to include and have other tables refer to an initial summary table for data definitions, such as when there is a large set of variables to explain.) Omit marginally useful information about variables such as their Compustat item numbers. (Such information can be mentioned once in the data section of the paper.) Write the explanatory paragraph more tightly than a typical paragraph in the body of a paper. Include little or nothing about the motivation for presenting the information or the conclusions that one can draw; in other words, focus on what, not why. Consider including a summary of the sample beyond what is in the title if it would aid the reader, especially in the first table to use the sample, but omit minute details and play-by-play accounts of the sample-construction process. This is not the place to list every detail needed to replicate the results (the body of the paper is, within reason). Do not repeat the table number or title. Do not list variable names, names of statistics and such when there is no need to further describe them and they are identified in the table title or in row, column or panel headings. Avoid Some tables need no explanatory paragraph, some need only one sentence and some need several sentences. Avoid mysterious variable names. There is no need to limit variable names to just a few characters or to a single word. Too many papers use such cryptic row headings in regression tables that one wonders if the author still encodes computer programs using a keypunch machine. (The editors have fond memories of doing so.) All column headings, row headings and so on should have only the first word capitalized (except for proper nouns and acronyms, of course). Panels: Table rows may be grouped into panels. Please consider dividing a table into more than one table instead of using long panels. If a table does incorporate panels, the panel headings should be left-justified with panel designators and initial capitals in the panel names. Panels should be designated alphabetically, e.g., Panel A: Causes of bankruptcy. In the text, refer to "Table 3, Panel A", or "Panel A of Table 3". A compressed notation such as "Table 3A" is unacceptable, whether in a table caption or in a text reference to the panel. Panels must not be further subdivided. Panels must appear in horizontal sequence, never side by side. Tables must not have footnotes, except for the significance asterisk legend when used. Example of table caption consisting of table number line, title and explanatory paragraph:
Statistical significance, when identified explicitly in a table, must follow these rules:
Unpublished appendices:A submitted paper may include an appendix, marked as not proposed for publication, containing material potentially important for the review process but that would not be of sufficient interest to most readers of the paper to justify publication. Such an appendix must be included in the same PDF as the rest of the paper for review purposes. StyleNumbersZero through ten are spelled out, then 11, 12, ... are written as numerals, with exceptions such as one hundred, 36 thousand, 4.2 million (not millions), etc. and first words of sentences. Use commas in numbers with more than three digits (e.g., 1,234 vs. 1234). Percents appear as percent sign, e.g., 12.34%, not written out, e.g., 12.34 percent. Precede all decimals with a zero, e.g., 0.1234, not .1234. A number or percent at the beginning of a sentence is spelled out, e.g., "Forty-two of these firms are in the full sample ..." but FR prefers that authors use a different construction (e.g., "The full sample includes 42 of these firms ..."). Months and years are written without commas or apostrophes (e.g., 1980s, January 1990). A full date takes a comma after the day but not after the year (e.g., "The January 13, 1990 announcement date ..."). Centuries can be expressed with numerals, e.g., 20th century. Hyphenate compound fractions, e.g., 1-1/2. Spell out simple fractions, e.g., one third. Levels of significance in the body of the paper should follow the format described. Displayed formulae should be numbered consecutively throughout the manuscript as (1), (2), etc., in the right-hand margin of the page. In cases where the derivation of formulae has been abbreviated, it helps the referees if the full derivation is presented on a separate sheet. (This sheet will not be published.) Use 12-pt regular, not boldface (BF). No punctuation in description list. Columns of numbers should be decimal-aligned with a consistent number of decimal places. Overused Words and Word UsageOverused and Frequently Misused Terms"And/or" is less readable than "or" which does not rule out both. We do not permit and/or in accepted papers. "As such" is not a general substitute for "therefore" in U.S. English. It only makes sense when an antecedent noun or construct, typically the predicate complement, could replace "such." (A black cat is a dark object. As such, it does not reflect much light. The predicate complement "a dark object" could replace "such", so this usage is okay.) "By:" Too many authors make awkward use of constructions like "We do our thing by doing something else." This is fine when "something else" is the complete or predominant method of accomplishing "our thing": "We decorate the walls by painting them." When describing only one salient detail of the method, it is better to omit "by": "Ikenberry and Ramnath (2002) conduct their tests using daily data." "Extant" means existing but is best reserved for the specific sense of "not yet extinct". Financial economists have picked up the bad habit of writing "extant literature" meaninglessly (when it is not being contrasted with non-existent literature.) Substitute "the literature," "other studies", "research," "researchers" etc. If it is necessary to make explicit the difference between references to the literature and the current paper, which is often perfectly obvious anyway, one can use "previous" or "existing" but, please, not "extant." "Factor loading" is a term from factor analysis and should be kept there. Please use conventional terms for regression coefficients, e.g., "regression coefficients." "Framework" is overused, often as a vaguely grandiose replacement for "model." We ask that this word be replaced with something simpler and more direct. "Impact" should be used sparingly to mean an unusually strong or surprising effect. "In terms of" is overused and tends to be a red flag for complex, confusing sentence structure. If you find yourself using it, try to rewrite the sentence to make it simpler and clearer. "In this paper" is fine as a transition from discussing other studies to discussing the current paper's contribution, and is not overused that way. However, the phrase is a waste of words in the abstract, and completely useless as the first words of the introduction or conclusion. It should never appear in those positions. "Interestingly" is overused to the point that we ask authors not to use it. Try to make everything in your paper interesting (or leave out that which is of no interest), then let the reader judge whether you succeeded. If you mean "surprisingly" or "unexpectedly", say so and make sure it's clear why. "It is important to note that": Strike this phrase and start the sentence with the next word. "Level of" should appear only when it serves a necessary function, for example to distinguish between the level and the corresponding rate of change that the author also is discussing, and even then only if the risk of confusion is non-trivial. "Likely" is just fine when referring to likelihood or, less formally, probability. Too many authors misuse it in stating a testable hypothesis about some quantity other than probability or likelihood. Also see "significant." "Loading:" See factor loading above. "Methodology" means the analysis of method; a typical paper should refer to how the authors (or a cited work) conduct the formal analysis as method, methods, test, tests, model, technique, approach, research design, etc., not methodology. Potential replacements for "methodological" include technical and (for empirical methods) econometric. "Moreover:" Puh-lease! Only a nerd would think it's cool to use this word. Okay, only a nerd would edit an academic journal. But we still don't think this word is cool and we ask you to drop it. "Multivariate" is appropriate for methods that explicitly model the joint probability distribution of more than one variable of primary interest. Please use "multiple" for a regression of one dependent variable on multiple regressors. The classical multiple regression method assumes nonstochastic regressors and therefore is no more multivariate than a simple linear regression. ("Variate" means random variable.) "Note that" or "Notice that" are useless and distracting in a research paper. Please leave them out and write in a dynamic manner that tends to stimulate greater attention to what you consider to be the more important points. "Not reported" makes no sense when the author then proceeds to report the thing. Please use "not reported in detail", "not in a table", "not tabulated", etc. "Posit" and "postulate", in theoretically rigorous disciplines such as financial economics, are best reserved for referring to axioms or assertions in formal theoretical modeling (including work being cited). We ask that authors not use these words in any other context. Between people, and perhaps concepts, there are "relationships." Between mathematical terms there are "relations." "Significant" in the statistical sense sometimes is used, wrongly, in stating a hypothesis. It is correct to state that one tests a hypothesis that, for example, a parameter is equal to zero; or stated as the alternative hypothesis, that the parameter is different from zero. It is incorrect to state that one tests whether a parameter is significantly different from zero. Statistical significance is an indication of confidence in the inference from the test; it is not part of the hypothesis and it is not something that one tests. Also see "likely." "Study on" (where study is a noun) isn't wrong, but in U.S. English, "study of" is preferred (but it's always "paper on"). "Stylized" is overused and should be avoided ("conventional" is a possible substitute.) "Then": If the order of certain sequential research actions, such as the application of sample screening criteria or the calculation of a particular estimate, is obvious, well-known, unimportant or irrelevant, as it usually is, then first-we-did-this-then-we-did-that wastes words and makes for less dynamic prose. Just describe what you do without adding extra verbiage related to order. "This" without a clear antecedent in the same paragraph should be avoided (as in "This implies ... ", "This shows ...", etc.). In general, "this" and "these" are overused. "The result implies" is easier on the reader than "this result implies." "Try": As Yoda says, "do or do not, there is no try." This is very applicable to research questions. The authors either investigate a question or not. Don't use "try" in describing your own or cited authors' efforts; just state what the research does or omits doing as needed. If you want to state an intended purpose while leaving it to the reader to judge whether you succeed, you can write "the purpose is to" or words to that effect. "Venue" is primarily a legal and entertainment term; better choices are "location" or "site." "Whereas" tends to be a bit overused and occasionally misused. It's fine if used correctly and not constantly. Preferred Usage"A" or "An", and "The": Writers who come from areas where English either is not dominant or has a strong local flavor sometimes make choices to use or not use indefinite and definite articles that seem awkward to most native English speakers. An article is usually needed between a preposition and a singular direct object, for example, "Stock returns come from the CRSP database." When the direct object denotes a concept or category rather than a specific item, use an indefinite article or no article, for example, "Smith (2009) shows that the result is driven by self-selection bias" — add the article "a" after "by" if you like, but usually not "the." This can be tricky; if the noun can denote one specific thing, it is typically best to make it plural to refer to the concept, category or class of such things, unless it is a collective noun as discussed in the paragraph below. "Research," "evidence" and "analysis" are primarily collective nouns in U.S. English, that is, the singular form encompasses multiple instances of the class. Therefore, the plural nouns "researches," "evidences" and "analyses" should not be used in FR articles. The first two will be perceived as erroneous, or at least archaic, quaint or confusing, by most U.S.-based readers. The word "data" is plural and takes a plural verb, as in "The data come from ...", not "The data comes from ..." "That" and "which" are not interchangeable, especially from a style point of view. "That" introduces a restrictive clause (not delineated by commas), e.g., "The book that I read last night was excellent ...", i.e., as opposed to the book that I read last week. "Which" introduces a nonrestrictive clause (either descriptive or parenthetical and frequently delineated by commas), e.g., "The book, which I read last night, was excellent ...", i.e., it happens that I read it last night but the main point is that the book was excellent. If putting a comma before "which" in a sentence changes the meaning of the sentence, then "which" has probably been used incorrectly. The verb "may" is weak and often can be omitted or replaced by "could" or "can," particularly in sentences in which some doubt is already being expressed. Avoiding "may" can permit additional simplification of the sentence; for example: "A board with an outside chair may be better able to maximize shareholder wealth if it is less influenced by managers' personal interests ..." would read better as "A board with an outside chair can better maximize shareholder wealth if it is less influenced by managers' personal interests .." Replace "in order to" with "to." FR style is not to use "document" as a verb; please replace with a simpler word such as "show". Spelling and PunctuationU.S. spelling is required for the final versions of accepted papers. For example: favor, behavior, practice and maximize (instead of favour, behaviour, practise and maximise) Note the spelling of "heteroskedasticity" (i.e., not "heteroscedasticity"). "Database" is one word, but "data set" should be two words. "Treasury bill" can also be written "T-bill." AcronymsSpell out first occurrence with acronym in parentheses immediately following. EXCEPTIONS: CRSP, NYSE, and Amex. Nasdaq originated as an acronym but was changed to simply a name long ago; therefore, it should not be in all caps unless making a historical reference to the origins of the market of that name. (The market sometimes spells its own name in all caps, but this is a corporate identity choice, not an indication that the name stands for anything. FR style, like that of the Wall Street Journal, requires Nasdaq.) Use consistent punctuation throughout the submitted paper. Avoid dashes, use semicolons sparingly, and always use the serial comma (e.g., "operating efficiency, debt capacity, and taxes" not "operating efficiency, debt capacity and taxes"). To make the typesetters' job easier, keep hyphenated words all on one line and avoid dividing a word at the end of a line. Prefixes (e.g., non, co, sub, under, over, etc.) are generally closed up (e.g., nonzero, coinsurance, overinvestment, intraday) although "pre-event" and "post-event" are preferred. Exempli gratia (e.g.) should always be abbreviated and followed with a comma: e.g., this is the format to use. Bold Face [BF], Italics [itals], etc.
Hyphenation [Hyph]Hyphens are used for true compound modifiers before the noun (e.g., after-tax income, risk-free rate, two-day return, three-digit SIC code, value-weighted index, stock-price reaction). Do not hyphenate compound modifiers that are adverbs ending in "ly" (e.g., previously acquired subsidiary, equally weighted index, publicly traded stock). "Non" words can and often should be hyphenated if they're "made up" words, e.g., non-tender offer. Quotation marks A single mark (') is an apostrophe whereas a double mark like " is a quotation mark. Do not use apostrophes as quotation marks except when quoting a passage that includes quotation marks, a practice that usually should be avoided in finance research papers. For a quote within a quote, use " for the outer pair (around the passage you are quoting) and ' for the inner pair (replacing the quotation marks in the original). Parentheses Never nest multiple sets of parentheses inside one another in the text. When a citation appears in a parenthesized sentence, set off the year with a comma instead of a second pair of parentheses. Separate serial citations that appear inside the same parentheses with semicolons. Present, Past and Future TenseCite the literature and report results in the present tense only. Use past tense to refer to historical events. (For example: On July 20, 2004, Microsoft Corp. announced the largest single special dividend and the largest repurchase program in history.) Avoid future tense. Use present tense, in the subjunctive if appropriate, to describe hypothesized relations. Style Glossary[ A,B,C | D,E,F | G,H,I | J,K,L | M,N,O | P,Q,R | S,T,U | V,W,X,Y,Z ] Accounting performance is not an acceptable term unless you mean how a firm's accounting system performs. Financial economists usually mean operating performance. Administrate should be replaced by administer. American depositary receipts/ADRs. From WSJ 5/11/98:
advisOrs (not advisErs) Altman's Z-score Amex (i.e., not all caps) is the abbreviation for American Stock Exchange, per the exchange's own usage. announce is what the party responsible for the action does; news outlets report antitakeover a priori is italicized but it is not hyphenated unless it's used as a compound adjective Arbitrage Pricing Theory (APT) AAR = average abnormal return available from the authors on request: as opposed to available from the authors whether we want it or not? "On request" is redundant. BBO = best bid or offer BILLION (must denote 109, not 1012 as is traditional in European English!) is NEVER plural. Same thing with MILLION and so on. Bera-Jarque's test bioengineering (one word, no hyphen) Black-Scholes options-pricing formula blockholders (1 word) "Broad Tape" Dow Jones News Wire/DJNW buy-and-hold abnormal return (BHAR) bylaws NOT HYPHENATED capital-asset pricing model (CAPM) HYPHENATED carryforwards/backs (not hyphenated) cash flow (two words) Center for Research in Security Prices (CRSP) U.S. Stock Databases (not "tapes") Cochrane-Orcutt procedure. Note that there is an e on the end of Cochrane, contrary to the spelling seen in some papers and statistical software. The procedure has been so well known for so long that the original article isn't commonly cited in journal articles, but for anyone interested, the citation is: Cochrane, D. and G. H. Orcutt, 1949. “Application of Least-Squares Regressions to Relationships Containing Autocorrelated Error Terms,” Journal of the American Statistical Association 44, 32-61. column: in-text ref, use #, e.g., "... is illustrated in column 5 ..." compare to -- if likening, compare with if setting side by side and looking at differences Composite (init cap OK when referring to NYSE) Compustat CIS Compustat Industry Segment (no need for itals) This is official S&P spelling. S&P Library: 212.208.8514. Fax 514.7016 Compustat database (not "tapes") counterargument Cournot competition (quantity setting) cross-holdings AND cross-shareholders cum day, ex day (no hyph) cumulative abnormal returns (CARs) CUSIP (not "number"; it has been alphanumeric for about a quarter century) cut-off as a noun; cut off in verb form Return to the top of the glossary.
data -- always a plural word data set (2 words) debt-to-value/equity ratio (hyphenate all such compound modifiers) decision making: NOT HYPHENATED when used as a noun (e.g., managerial decision making), HYPHENATED when used as a compound modifier (e.g., decision-making process) defined contribution pension plan: 401(k) dependent: dictionaries allow dependant as a variant spelling, but FR prefers dependent deviation: "two-standard-deviation" should be hyphenated different is often redundant when preceded by a number; "five samples" is easier to read than "five different samples." Also, please avoid the bad habit of using "different" to mean "several". In other words, use "different" only to communicate that one thing isn't the same as or similar to another. document Do not use as a verb. Use "show" or "report." Dow Jones News Wire Dow Jones Industrial Average dummy variable: omit "and zero otherwise"; readers of a finance journal know what a dummy variable is e.g., followed by comma. Do not italicize. "Empirical test" typically is redundant in a research context and can shortened to test. "Ensure" except when referring to "Insurance" and the protection it provides ESOPs (Employee Stock Ownership Plan) ERISA et al. is an abbreviation (needing a period at the end) of the Latin et alii (masculine), et aliae (feminine), or et alia (neuter), meaning “and [the] others.” (Note that the first word of the phrase, et, isn't abbreviated and so does not take a period.) Eventus (not all caps) ex ante and ex post are hyphenated only when used as compound adjectives ex day (no hyph) extant means not yet extinct; "extant literature" makes no sense Financial Review: Please cite as The Financial Review firm -- in general, use only to refer to law firms, small family-owned businesses, etc. -- retain if author uses it almost exclusively throughout the article fixed-beta CAPM forgo means give up or do without forego means to go before fraction: often this should be "percentage" -- unless author really is referring to a fraction such as 2/3, 1/2, or ticks framework tends to be used too much and too vaguely in finance research papers. Use model, method, theory etc. freeze-out law (Delaware, 1988) Return to the top of the glossary. GARCH (generalized autoregressive heteroskedasticity) models Generalized Method of Moments (GMM) group-based incentive plans heterosKedasticity (i.e., use the American spelling; not the European spelling with a "c") "Hewlett-Packard" IS hyphenated high-q firm IBES Institutional Brokers Estimate System (no slash marks) i.e., followed by comma USE REGULAR FONT, NOT ITALICS interestingly: as opposed to the rest of your paper? Seriously, whether it's your own results or others you are citing, it's better to write in such a way that the reader sees why it's a particularly striking point without your having to say so. Investment Responsibility Research Center (IRRC) ITSA (Insider Trading Sanctions Act of 1984) ITSFEA (Insider Trading and Securities Fraud Enforcement Act of 1988) Return to the top of the glossary. larger in magnitude: as opposed to larger but not in magnitude? larger will do Lexis-Nexis database liquid yield option note (LYON) Ljung-Box test statistic log is acceptable for logarithm in the text, but in math expressions use ln to denote the natural log long run: hyphenate only when used as a modifier, e.g. long-run returns exist only in the long run-q firm Return to the top of the glossary. MVBV market-to-book value may -- AVOID the verb "may." Weak. Use only in cases where it's absolutely correct Mergerstat Review, W.T. Grimm Methodology should be reserved for denoting the investigation of methods or laborious development of new methods. FR style is not to use the word to denote the body of methods employed in an article; use simpler words like methods, approach, tests, etc. instead. Nasdaq has not been an acronym for many years. Although the market resumed identifying itself as NASDAQ sometime during the first decade of the 21st century, this is because they made a style choice, not because the name stands for anything as it did in the 1970s. The Financial Review, like the Wall Street Journal, now uses the form Nasdaq. nonfinancial (NOT hyphenated) NPV net present value On ... is usually preferred to "upon" . May not be the first word of a paper's title. option-pricing model -- HYPHENATE Return to the top of the glossary. p -value -- the P is italicized PACAP = Pacific Basin Capital Markets Research Center Pareto-optimal (hyphenated when used as a compound adjective) Pension Benefit Guaranty Corp (PBGC) percentage points: example: from 8.7% to 15.7% = seven additional percentage points, not an additional 7% pre- and post-dividend, buyout, etc. okay to hyphenate, BUT do not hyphenate pretax premiums, rather than premia profit-sharing plan (hyphenated when used as a compound adjective) q as in Tobin's q ratio high-q firm, low-q firm ratio, market-to-book (hyphenated) relations between variables, relationships between people return on the market value (ROMV) reveal -- save this for what strippers do. Use "show" or "demonstrate" instead. run-up Return to the top of the glossary. SDC -- Cite in first mention as Thomson Financial SDC Platinum (SDC), and refer to as SDC thereafter. Thomson acquired Securities Data Company several years ago. SEC -- U.S. Securities and Exchange Commission is the full name (not Securities Exchange Commission). Omit U.S. if the paper uses only U.S. data. spin-off stand-alone Standard & Poor's (ampersand [&] okay in this exceptional case) Standard & Poor's 500-stock index SmallCap 600 standard deviation -- hyphenated if used as compound adjective Standard Industry Classification (SIC) When an author defines the SICs by name (e.g., Textile Mill Products, Pulp and Paper, etc.), use initials and all-capitals standard unexpected earnings (SUE) stock price based compensation -- DO NOT HYPHENATE "STOCK PRICE" compounds stylized Replace with substitutes such as "abstract", "conventional", "generally accepted" t-statistic -- the T is lower case and italicized; in table column headings, just t tax-loss carry-forward 10-K (with a hyphen) tercile or tertile: A group containing a third of a population. Both are correct, but we ask that you use tercile. time period should be simply period (unless the reader would be likely to mistake it for another kind of period) time series singular or plural; e.g. one time series shows autocorrelation, two time series show autocorrelation Tobin's q ratio or Tobin's Q tobit (no initial capitals) specification tradeoff (noun) and trade off (verb) Trades, Orders, Reports, and Quotes (TORQ) database transaction costs, not transactions costs Treasury bill, T bill -- DO NOT HYPHENATE unless it's used as a compound modifier, e.g., T-bill rate UPON usually should be replaced by ON Return to the top of the glossary. venue is a legalism. Use "location" or "site" instead. vice president -- no hyphen; use caps only when referring to specific person, e.g., "Vice President Jones," "the vice president of the company" Wal-Mart Wilcoxon signed-rank tests writedown (1 word, no hyphen) Z-score, Altman's zero always SPELLED in text to avoid any possible confusion Return to the top of the glossary. Copyright © 1997–2008 Eastern Finance Association |