我把四本关于论文写作的书做成了一个skill，以下是中间过程，保留了大量的引用细节，个人觉着是非常值得保留并阅读的内容。我把这些中间过程做成了Marp，供以后上课使用。参考的书包括：

• Dudenhefer, Paul. 2009. A Guide to Writing in Economics. Duke University.
• McCloskey, Deirdre N. 2000. Economical Writing. 2nd ed. Waveland Press.
• Nikolov, Plamen. 2013. “Writing Tips For Economics Research Papers.” Harvard University.
• Wallwork, Adrian. 2011. English for Writing Research Papers. Springer.

Economics Research Paper Writing Guide

Generating clear, persuasive, and publication-ready economics research papers in English.

Based on: Nikolov (2013), Wallwork (2011), Dudenhefer (2009), and McCloskey (2000)

Table of Contents

1. Core Principles (Clarity, Conciseness, Audience)
2. Sentence-Level Style (Active verbs, Short subjects, Old→New, Nominalizations, Vague words, Elegant variation, This/That, Concrete)
3. Paragraph-Level Style (Points, Coherence, Tone)
4. Common Mistakes (Boilerplate, False precision, Jargon)
5. Paper Structure (Inverted pyramid, Empirical, Theoretical)
6. Writing the Abstract
7. Writing the Introduction (Four-move structure)
8. Writing the Literature Review
9. Writing an Empirical Paper (Data, Model, Identification, Results)
10. Writing a Theoretical Paper (Model, Propositions, Proofs)
11. Designing Tables and Figures
12. Writing the Conclusion
13. Citations and References
14. Final Quality Checklist

Part 1: Core Principles

1.1 Clarity Above All

Clarity is the single most important criterion for economics writing. As McCloskey (2000, p. 12) quotes Quintilian: “one ought to take care to write not merely so that the reader can understand but so that he cannot possibly misunderstand.”

Clarity is a social matter: if the reader finds a passage unclear, it is unclear by definition (McCloskey 2000, p. 12).

Nikolov (2013) emphasizes that clarity trumps snappiness: “Do not worry about being snappy if you are being clear.” Readers tolerate a less-than-scintillating delivery if it serves clarity.

Rule: Every sentence must be clear on first reading. If a sentence requires re-reading, revise it.

1.2 Be Concise

All four sources agree: cut relentlessly.

McCloskey (2000, p. 70) quotes Sydney Smith: “In composing, as a general rule, run your pen through every other word you have written.”

Nikolov (2013) puts it even more bluntly: “Cut, cut, and then cut again.”

Wallwork (2011, p. 84) notes that shorter writing reduces errors and increases readability.

Dudenhefer (2009, p. 12) recommends Joseph Williams’s principle: “Express key actions as verbs” rather than burying them in nominalizations (e.g., “conduct a review” → “review”).

Rule: After generating a draft, remove 20% of the words without losing meaning. Replace every “is,” “there is,” and “it is” with active verbs where possible.

1.3 Audience Awareness

You are writing for busy, impatient readers who skim.

Nikolov (2013) warns: “No reader will ever read the whole thing from start to finish. Readers skim. You have to make it easy for them to skim.”

Dudenhefer (2009, p. 34) recommends choosing an implied reader—such as a smart college roommate who has never taken economics—and sticking with that reader throughout.

Wallwork (2011, p. 13) emphasizes that referees review papers in their own time with no financial reward, so “do everything you can to make the referee’s work easier and more pleasurable—clear English, clear layout, clear tables etc.”

Rule: Define the implied reader explicitly before generating content. What does this reader already know? What do they need to know? What would surprise or puzzle them?

Part 2: Sentence-Level Style

2.1 Use Active Verbs

Passive voice weakens prose.

Nikolov (2013): “The passive voice is avoided by good writers.”

McCloskey (2000, p. 70): “Use active verbs: not ‘active verbs should be used,’ which is cowardice, hiding the user in the passive voice.”

Wallwork (2011, p. 24) shows that active sentences are shorter and clearer.

Example:

• Passive: “The passive voice is avoided by good writers.”
• Active: “Good writers avoid the passive voice.”

Rule: For every sentence, identify who is doing what. Make the doer the subject and the action the verb.

2.2 Keep Subjects Short

Dudenhefer (2009, p. 12), following Williams (2006), states: “Readers like to get past the subject to the verb as quickly as possible.” Long grammatical subjects slow the reader.

Example:

• Long subject: A full explanation of why the model cannot accommodate this particular case of omitted variable bias is given in the appendix.
• Short subject: The appendix explains in full why the model cannot accommodate this particular case of omitted variable bias.

Rule: If a sentence’s subject exceeds 10 words, revise it. Move the long phrase to after the verb.

2.3 Old Information First, New Information Last

Dudenhefer (2009, p. 12), following Gopen (2004), emphasizes two principles:

1. Begin sentences with “old” information (what the reader already knows).
2. End sentences with “new” information (what you want to emphasize).

This creates flow and helps the reader follow your argument.

Example:

• Weak: An effective way to write sentences that “flow” is to use the rhetorical device known as conduplicatio. To repeat a key word or phrase from a preceding sentence, especially when the word or phrase comes at the end of the preceding sentence, is to use conduplicatio.
• Strong: An effective way to write sentences that “flow” is to use the rhetorical device known as conduplicatio. Conduplicatio repeats a key word or phrase from a preceding sentence, especially when the word or phrase comes at the end of the preceding sentence.

Rule: In every sentence, identify what is old and what is new. Put the old at the beginning and the new at the end.

2.4 Avoid Nominalizations

Nominalizations are noun forms of verbs (e.g., “analysis” from “analyze,” “assumption” from “assume”). They sap vigor from prose (Dudenhefer 2009, p. 12; McCloskey 2000, p. 71).

Examples:

• Economists made attempts to define full employment. → Economists attempted to define full employment.
• The model makes the assumption that people engage in utility maximization. → The model assumes that people maximize their utility.

Rule: Search for words ending in “-tion,” “-ment,” “-ence,” and “-ity.” Convert them to verbs where possible.

2.5 Avoid Vague Words and Phrases

McCloskey (2000, pp. 72–77) provides a blacklist of words that signal bad writing:

Nikolov (2013) adds: avoid “of course,” “clearly,” “obviously,” and “very.”

Rule: Scan every sentence for words in the blacklist. Delete or replace them.

2.6 Avoid “Elegant Variation”

Do not use different words to mean the same thing.

McCloskey (2000, p. 56): “Use one word to mean one thing. Get your words and things lined up and keep them that way.”

Wallwork (2011, p. 103) warns against “monologophobia—the constant search for synonyms.”

Example:

• Bad: “industrialization,” “growing structural differentiation,” “economic and social development,” “social and economic development,” “development,” “economic growth,” “growth,” and “revolutionized means of production” (all meaning the same thing).
• Good: Choose one term and stick with it.

Rule: If you introduce a technical term, use that exact term consistently. Do not alternate between “consumer,” “household,” and “agent.”

2.7 Avoid This/That/These/Those as Pronouns

McCloskey (2000, p. 85): “Circle the ‘this’ and ‘these’ in your draft: you’ll be surprised at their number.” These words force the reader to look backward, interrupting the forward flow.

Example:

• Weak: This suggests that…
• Strong: This result suggests that…

Rule: Query every “this” or “these.” If it does not have a noun immediately following it, add one.

2.8 Be Concrete and Plain

McCloskey (2000, pp. 78–80) insists: “A singular word is more concrete than a plural. Definiteness is concrete. Prefer Pepperidge Farm to bread, bread to widgets, and widgets to X.”

Avoid five-dollar Latinate words when a plain Anglo-Saxon word will do.

Example:

• Bad: “The integrative consequences of growing structural differentiation.”
• Good: “The need for others that someone feels when he buys rather than bakes his bread.”

Nikolov (2013): “Keep sentences short. Short words are better than long words. Monosyllabic words are best.”

Rule: For every abstract term, ask: Can I give a specific example? For every long word, ask: Is there a shorter, plainer word?

Part 3: Paragraph-Level Style

3.1 Paragraphs Should Have Points

Each paragraph should discuss one topic and one topic only.

McCloskey (2000, p. 44): “The paragraph should be a more or less complete discussion of one topic.” If a paragraph is too long, readers will skip it. If too short, the writing feels breathless.

Rule: Every paragraph should pass the “one-sentence summary” test: Can you summarize its point in one sentence? If not, split it.

3.2 Coherence Through Repetition

McCloskey (2000, p. 50) advocates coherence through repetition of key words, not through clanking conjunctions (“however,” “furthermore,” “therefore”). English achieves coherence by repetition, not by signal.

Example:

• Weak (over-signaled): However, not only did good Latin prose have this feature, but also Greek had it. Therefore, many Ciceronian adverbs are untranslatable. Yet, to be sure, the impulse to coherence is commendable. But on the other hand…
• Strong (repetition): Good Latin prose repeated key words. Greek prose did too. English prose should repeat key words as well. Repetition brings clarity.

Rule: In every paragraph, ensure that each sentence is linked to the previous one by repeating a key word or using a pronoun that clearly refers back.

3.3 Control Your Tone

McCloskey (2000, pp. 39–43) warns against adopting a pompous “Scientist” persona. The best tone is “the Reasonable and Modest Journeyman.”

Avoid invective (“This is pure nonsense”)—it makes readers sympathize with your target, not with you (Smith 1759, I.ii.3.4, cited in McCloskey 2000, p. 41).

Nikolov (2013) adds: avoid adjectives and verbs that are overly dramatic (“the results shatter our expectations”).

Rule: When generating text, adopt the persona of a reasonable colleague explaining something to a smart friend. Avoid the “Scientist” or “Enthusiastic Student” voice.

Part 4: Common Mistakes to Avoid

4.1 Boilerplate

McCloskey (2000, pp. 35–38) identifies several forms of boilerplate:

• Excessive introduction and summarization
• “This paper…” openings
• Table-of-contents paragraphs (“The outline of this paper is as follows”)
• Background padding

Nikolov (2013): “A good paper is not a travelogue of your search process. The reader doesn’t care how you came to figure out the right answer.”

Rule: Cut the first paragraph of every section and see if the section still makes sense. If yes, leave it cut.

4.2 False Precision

Nikolov (2013): “Don’t report 1.23456789.” Use standard errors to guide rounding: report coefficients to the same decimal place as the standard error, plus one additional decimal place.

Rule: For every number in the paper, ask: Is this level of precision justified by the standard error? Round aggressively.

4.3 Jargon and Acronyms

McCloskey (2000, p. 83): “Resist [acronyms], and remember that even expert mathematicians do not think in symbols.”

Nikolov (2013): “Never make up your own acronyms.”

Rule: Limit acronyms to widely known ones (GDP, OLS). For everything else, use the full term or “it.”

Part 5: Paper Structure Overview

5.1 The Inverted Pyramid (Newspaper Style)

Economics papers should be organized like newspaper articles, not like jokes or mystery novels (Nikolov 2013). In a joke, the punchline comes at the end; in a newspaper article, the most important information comes first.

Why this matters: Your readers are busy and impatient. “No reader will ever read the whole thing from start to finish. Readers skim. You have to make it easy for them to skim” (Nikolov 2013). Most readers want to know your basic result immediately.

Rule: State the central contribution in the first paragraph of the introduction. Do not “clear your throat” with long motivation, policy importance, or literature surveys before saying what the paper does.

5.2 Standard Structure: Empirical Paper

Empirical economics papers typically follow this order (Dudenhefer 2009, p. 35; Nikolov 2013):

1. Title
2. Abstract
3. Introduction
4. Literature Review (sometimes integrated)
5. Data
6. Model / Identification Strategy
7. Results
8. Discussion (sometimes integrated)
9. Conclusion
10. References
11. Appendices (if needed)

Nikolov (2013) advises organizing the paper “from the inside out”: the model and empirical work form the core, with the introduction and conclusion framing them last.

Rule: Generate the Data or Methods section first if the introduction is not immediately clear. These sections are typically the most straightforward to write.

5.3 Standard Structure: Theoretical Paper

Theoretical papers have a different balance (Dudenhefer 2009, p. 30):

1. Title
2. Abstract
3. Introduction
4. Literature Review (often brief)
5. Model (extensive, often spanning multiple sections)
6. Propositions, Proofs, Theorems, Lemmas
7. Discussion / Extensions
8. Conclusion
9. References

Unlike empirical papers, theoretical papers may contain as much mathematics as prose. They are identified by an abundance of “propositions,” “proofs,” “theorems,” and “lemmas” and by an absence of data.

Rule: In theoretical papers, introduce the model by “moving from infrastructure to superstructure” (Thomson 2001, cited in Dudenhefer 2009, p. 37). Introduce and describe each actor category separately before bringing them together.

5.4 Length Guidelines

For a 15–20 page term paper, Nikolov (2013) suggests the following approximate lengths (single-spaced):

Rule: If a section exceeds its recommended length, determine whether material belongs in an appendix or footnote, or whether the section is repeating itself.

5.5 The “So What?” Test

Every section of your paper must answer the reader’s implicit question: “So what?” (McCloskey 2000, p. 29). Why does this matter? Why should I care?

Rule: For every paragraph, verify that it answers “So what?” If a paragraph does not explain why it matters, delete it or revise it to make its importance explicit.

5.6 Appendices: The Overflow Valve

Appendices are a great tool for material that supports your argument but would disrupt the flow of the main text (Nikolov 2013). This includes:

• Robustness checks
• Additional tables and figures
• General versions of the model
• Detailed proofs
• Data descriptions

Rule: Once a robustness check is verified not to change the main result, move it to an appendix. Summarize the findings in the main text.

5.7 Opening Hook

Never start a paper with “This paper…” (McCloskey 2000, p. 36). Instead, use a hook that brings the reader from where they presumably stand in knowledge to where your paper will take them.

Good examples:

• “Every economist knows by now that monopoly does not much reduce income. Every economist appears to be mistaken.” (McCloskey 2000, p. 36)
• “More immigrants entered the United States during the past decade than in any comparable period since the 1920s.” (Butcher and Card 1991, cited in Dudenhefer 2009, p. 46)

Rule: Generate at least three alternative opening sentences. Select the one that most directly challenges the reader’s intuition or presents a striking fact.

Part 6: Writing the Abstract

6.1 What Is an Abstract?

An abstract is a paragraph-long condensation of the main elements of your paper. It is the most-read part of your paper—many readers will read only the abstract and perhaps the conclusion (Wallwork 2011, p. 177; Nikolov 2013).

Length: Usually 150 words or less (Dudenhefer 2009, p. 52). Some journals allow up to 250 words, but shorter is better.

When to write it: Last. “The abstract (the summary of the paper, which you will write at the very end once you have actual results)” (Nikolov 2013). Wallwork (2011, p. 179) agrees: write a rough draft early to orient your ideas, but finalize it after the paper is complete.

6.2 What to Include

An abstract should answer these questions (adapted from Dudenhefer 2009, p. 52; Wallwork 2011, p. 180):

1. What does the paper do? (purpose / research question)
2. How does it do it? (data / methodology)
3. What are the principal findings? (main results)
4. What are the implications? (contribution / significance)

Example (short, 58 words):

We selectively survey, unify and extend the literature on realized volatility of financial asset returns. Rather than focusing exclusively on characterizing the properties of realized volatility, we progress by examining economically interesting functions of realized volatility, namely realized betas for equity portfolios, relating them both to their underlying realized variance and covariance parts and to underlying macroeconomic fundamentals. (Anderson, Bollerslev, and Diebold 2005, cited in Dudenhefer 2009, p. 52)

6.3 What NOT to Include

Wallwork (2011, p. 191) and Nikolov (2013) warn against:

• Background material that belongs in the introduction
• Citations (unless absolutely necessary)
• Abbreviations and acronyms that are not standard
• Vague claims (“the results are discussed”)
• References to tables, figures, or sections of the paper
• Speculation about future research

Rule: If a sentence in the abstract does not directly answer one of the four questions above, delete it.

6.4 Tense and Style

• Use the present simple for established facts and your main findings: “We find that…”
• Use the past simple for your specific methodology: “We used data from…”
• Avoid the passive voice: “We estimate” rather than “An estimate is made”

Wallwork (2011, p. 186) notes that some journals prefer the impersonal style (“This paper finds…”) while others accept the personal style (“We find…”). Check your target journal’s guidelines.

Rule: Generate two versions of the abstract—one in active voice, one in passive—and select the clearer one.

6.5 The “Elevator Pitch” Test

McCloskey (2000, p. 29) recommends imagining you have 30 seconds to explain your paper to a colleague in an elevator. Your abstract should be that 30-second explanation in writing.

Rule: The abstract should be readable in 30–45 seconds (approximately 150 words or less).

Part 7: Writing the Introduction

7.1 The Four-Move Structure

Dudenhefer (2009, pp. 46–47), following Swales and Feak (1994), recommends thinking of your introduction as progressing through four “moves”:

Move 1: Establish a Research Territory

Introduce your subject and indicate its importance. Why should we care? Why is it important to economics or human welfare?

Move 2: Review the Literature

Place your own study in the context of other studies. Highlight what has been done and what remains unresolved.

Move 3: Establish a Niche

Identify a gap, problem, or deficiency in the current literature. This is the opportunity your study will address.

Move 4: Occupy the Niche

State your study’s purpose, what it will contribute, and your main findings. End with a brief outline of the paper’s structure (optional).

Important: These moves need not occur in strict order. Some papers begin with Move 4 (“This paper will…”), while others weave the moves together (Dudenhefer 2009, p. 47).

7.2 Start with Your Contribution

Nikolov (2013) is emphatic: “Do not start with philosophy… Do not start with ‘The finance literature has long been interested in x.’ Your paper must be interesting on its own, and not just because lots of other people wasted space on the subject.”

McCloskey (2000, p. 36) agrees: “Never start a paper with that all-purpose filler for the bankrupt imagination, ‘This paper…’” Instead, start with a hook.

Good openings (Dudenhefer 2009, pp. 46–47; Nikolov 2013):

• Stating what the paper does: “This paper develops a consumption-based model of asset pricing which integrates the real financial and monetary sectors of the economy.” (Boyle and Young 1988)
• Describing an economic problem: “How do increases in competition affect equilibrium bidding at auctions?” (Hong and Shum 2002)
• Providing relevant background: “For 20 years following 1949, average total fertility per woman in China hovered just above six children.” (McElroy and Yang 2000)
• Citing previous studies (most common): “Several recent papers have studied the univariate time-series process for U.S. GNP…” (West 1988)

Rule: Generate the introduction only after drafting the body of the paper. Make the first sentence directly state what the paper does and why it matters.

7.3 Challenge the Reader’s Intuition

Nikolov (2013) recommends: “As soon as you mention [your contribution], mention something unexpected about it! The reader will be much more motivated to read the rest of the paper if you challenge his or her intuition right from the get-go.”

Example:

Ever since Harrod’s (1948) discussion of hump savings, economists have recognized the importance of saving during working years for consumption during retirement. Hump-savings refers to the shape of an individual’s savings curve over time: low at the beginning, higher in the middle, lower at the end. This basic model is used throughout the paper and holds together all that follows. Feldstein (1974) goes on to argue that while the hump-savings model works well to explain most of the observed data, the effect of certain government policies on individual savings has never been measured empirically. (Nikolov 2013, paraphrasing Feldstein 1974)

Rule: After stating the research question, add one sentence that surprises the reader or contradicts common wisdom.

7.4 Length and Scope

• Undergraduate term paper (15–20 pages): Introduction should be at most 1–2 pages single-spaced (Nikolov 2013).
• Journal article: Typically 2–4 pages, depending on the journal.

Nikolov (2013): “Two pages is a good upper limit for the introduction.” Dudenhefer (2009, p. 46) suggests approximately one paragraph per five pages of essay as a rough guide.

What to cover (Nikolov 2013):

1. The question you address (state the hypothesis directly)
2. Why we should care (from an economic perspective, not just policy importance)
3. What your contribution is (new data? new model? new identification strategy?)
4. What your main results are (briefly)
5. How your findings differ from previous work

Rule: If the introduction exceeds two pages, cut background material that belongs in the literature review or data section.

7.5 Avoid Common Openings

Nikolov (2013) and McCloskey (2000, p. 36) list openings to avoid:

• “This paper investigates…”
• “Financial economists have long wondered if…”
• “The literature has long been interested in…”
• “It is well known that…”
• “In recent years, there has been growing interest in…”

These are “throat-clearing”—wasted space that delays your contribution.

Rule: Search the introduction for the phrases above. Replace each with a direct statement of what the paper does or a striking fact.

7.6 Ending the Introduction

The introduction should end by previewing the paper’s structure. However, this preview should be brief—one sentence or a short paragraph at most. Do not write a table-of-contents paragraph (McCloskey 2000, p. 37).

Good example:

In Section 2 we describe the data. Section 3 presents the model and identification strategy. Section 4 reports the results, and Section 5 concludes.

Bad example:

The outline of this paper is as follows. In Section 2, we review the literature. In Section 3, we describe our data. In Section 4, we present our model. In Section 5, we discuss our estimation strategy. In Section 6, we report our results. In Section 7, we conclude.

Rule: If the preview of the paper’s structure exceeds two sentences, it is too long. The structure should be obvious from the section headings.

7.7 Writing the Introduction Last

Dudenhefer (2009, p. 46) and Nikolov (2013) both recommend writing the introduction after the rest of the paper is complete.

“Once you have completed the discussion of your results, you can add a short conclusion summarizing what you have done. Then go back and write an introduction that provides a road-map for the reader” (Nikolov 2013).

Rule: After generating the body of the paper, distill the central contribution into one paragraph. That paragraph becomes the core of the introduction.

Part 8: Writing the Literature Review

8.1 What a Literature Review Is NOT

Dudenhefer (2009, p. 55) is clear: “A literature review is not just a description of a series of papers; it is not a mere catalog or annotated bibliography of papers written on a subject. A series of paragraphs, each recapping or summarizing a particular paper or set of papers, in no particular order, does not a literature review make.”

Nikolov (2013) adds: “Do not start your introduction with a page and a half of other literature. First, your readers are most interested in just figuring out what you do.”

8.2 What a Literature Review IS

A good literature review is “an account of previous research that is carefully constructed to tell a particular story” (Dudenhefer 2009, p. 55). The story is:

1. Here is what previous researchers have done.
2. Here is something unsatisfactory, incomplete, or troubling about that research.
3. Here is how I am going to redress it.

The review “turns” on a word or observation that signals a problem and a solution: however, although, in contrast, yet (Dudenhefer 2009, p. 55).

Examples of the “turn”:

• “Smith and Jones have done a wonderful job in their papers. However, they make one dubious assumption. In this paper, I will make another assumption, one that is more realistic or plausible.”
• “The existing empirical work has substantial limitations that the present study seeks to overcome” (Cohen and Einav 2003, cited in Dudenhefer 2009, p. 55).
• “Although there have been many studies which develop consistent estimators of the number of factors, the corresponding estimates… often considerably disagree. The purpose of this paper is to develop formal statistical tests” (Onatski 2009, cited in Dudenhefer 2009, p. 55).

8.3 Structure and Organization

A literature review should have a discernible principle of organization. Common approaches (Dudenhefer 2009, p. 55):

• Thematic: Group papers by methodology, data, or findings.
• Chronological: Trace the development of ideas over time.
• Methodological: Compare papers by their empirical strategies.

Example (thematic):

All four studies took a position on whether rural or urban development should be favored. Epstein and Joseph (2000) favor rural development. In contrast, Bhattarchya (2001) and Van Neer (2000) believe urban development is more important. Marshall (2003) concludes that it does not matter: either kind of development is equally beneficial. (Dudenhefer 2009, p. 55)

Rule: Before generating the review, organize the literature by author, question, data, method, finding, and limitation. Identify patterns and structure the review around those patterns.

8.4 What to Include

Nikolov (2013) recommends covering:

1. Previous research directly relevant to your paper (not every paper on the topic)
2. Papers that employ the same methods, analyze similar models, or use the same data
3. Your contribution in detail: Is it new data? A new model? A new identification strategy?
4. How your approach differs from and improves upon previous work

Dudenhefer (2009, p. 55) adds:

• Begin with comments about the body of research as a whole.
• Explain the merits and shortcomings of existing studies explicitly.
• Explain how your study will make a contribution.

Rule: For every paper cited, explicitly state how it relates to the present work—whether it supports, contradicts, or motivates the approach.

8.5 Generosity in Citations

Nikolov (2013): “Be generous in your citations. You do not have to say that everyone else did it all wrong for your approach and improvements to be interesting.”

Dudenhefer (2009, p. 55): A literature review is, “in a sense, a sales job. What is it selling? The value added by the present paper.”

Rule: Criticize previous work gently. Use phrases like: “Although Rodriguez (2001) and Dudley (2000) ask the right questions, their studies are hampered by data sets with an insufficient number of observations” (Dudenhefer 2009, p. 55).

8.6 Length and Placement

• Undergraduate term paper: About 2 single-spaced pages (Nikolov 2013)
• Journal article: Varies; may be a separate section or integrated into the introduction

Nikolov (2013): “Do not title your literature review section ‘Literature Review’! It is sophomoric. Integrate your discussion under a common thread… tell your readers what is in the section.” For example: “Gender Norms in India” rather than “Literature Review.”

Rule: If the paper has a separate literature review section, give it a descriptive title that signals the content.

8.7 Sources and Scope

Nikolov (2013) emphasizes that your literature review should consist mostly of economics papers, even if your topic is interdisciplinary. For interdisciplinary topics (e.g., health), it is fine to cite a couple of papers from other disciplines, but “mostly your work will be judged based on its relation to previous economics papers.”

Dudenhefer (2009, p. 23) recommends using:

• The New Palgrave Dictionary of Economics for overview entries
• Journal of Economic Literature (JEL) and Journal of Economic Perspectives (JEP) for literature surveys
• EconLit for searching economics-specific literature

Rule: Before generating citations, review encyclopedia entries (e.g., New Palgrave Dictionary of Economics) and survey articles from the Journal of Economic Literature or Journal of Economic Perspectives to map the landscape of the topic.

Part 9: Writing an Empirical Paper

9.1 Overview

An empirical paper tests a model with data to see how well the model represents reality—specifically, to what degree the model yields predictions consistent with observed behavior (Dudenhefer 2009, p. 29).

The standard sections are:

1. Data
2. Model / Identification Strategy
3. Results
4. Discussion (sometimes combined with Results)

Writing order: Dudenhefer (2009, p. 35) and Nikolov (2013) both recommend writing from the “inside out”: start with the model and empirical work, then write a narrative around selected tables and figures, and finally write the introduction and conclusion.

9.2 Data Section: What to Include

The data section should be approximately 1 single-spaced page (Nikolov 2013) and cover:

1. Name and source of the data: “This study uses data from the 1999 wave of the Panel Study of Income Dynamics” (Dudenhefer 2009, p. 36).
2. Period covered: What years? What frequency?
3. Type of data: Panel, cross-section, or time series? What is the unit of observation?
4. Number of observations
5. Limitations: Missing variables, missing observations, attrition, selection bias, small sample sizes
6. Descriptive statistics: A table with means, standard deviations, max, and min for key variables

9.3 Describing Data Sources

Dudenhefer (2009, p. 36) recommends:

• State the strengths and weaknesses of your data source.
• Compare your data to other sources used in the literature.
• Note any features that may affect results (overrepresentation, attrition, changes in collection methods).
• Explain any computations or adjustments you made.

Example (adapted from Dudenhefer 2009, p. 36):

This study uses data from the 1999 wave of the Panel Study of Income Dynamics (PSID). The PSID is a nationally representative longitudinal survey that began in 1968 and follows individuals and their descendants over time. The 1999 wave contains 7,406 households and 18,149 individuals. Our analysis sample is restricted to heads of household aged 25–64 who report positive earnings, yielding 4,892 observations.

9.4 Descriptive Statistics

Include a table with summary statistics. Nikolov (2013): “You should define each variable carefully and, if necessary, point out how the empirical measure deviates from its theoretical counterpart.”

Typical contents:

• Mean and standard deviation for continuous variables
• Percentages for categorical variables
• Histograms for skewed distributions

Rule: Before presenting the analysis, include a descriptive statistics table. This builds credibility and helps readers interpret the results.

9.5 Model Section: Describing Your Model

Economic models are abstract, simplified representations of an economy or decision-making process, expressed in words and mathematics (Dudenhefer 2009, p. 37).

How to present a model:

1. Begin verbally: Describe the economic agents, their objectives, and constraints before writing equations.
2. State the source: Did you construct the model yourself, or is it borrowed/adapted? “The model, like many in the new trade literature, is a variant on the monopolistic competition framework initially proposed by Dixit and Stiglitz” (Krugman 1991, cited in Dudenhefer 2009, p. 37).
3. Present equations: Use clear notation and define every variable immediately.
4. Explain assumptions: Lay out all assumptions and explain the intuition behind them.

9.6 Model Section: Example

Example (Krugman 1991, cited in Dudenhefer 2009, p. 37):

We consider a model of two regions. In this model there are assumed to be two kinds of production: agriculture, which is a constant-returns sector tied to the land, and manufactures, an increasing-returns sector that can be located in either region. All individuals in this economy are assumed to share a utility function of the form

U = C_M^μ C_A^(1-μ)

where C_A is consumption of the agricultural good and C_M is consumption of a manufactures aggregate.

9.7 Notation

Dudenhefer (2009, p. 37) advises:

• Follow standard notation in the literature, or make notation self-explanatory.
• Use subscripts, superscripts, and Greek letters liberally.
• Use specialized equation editors (Equation Editor, MathType, LaTeX).

Rule: After generating the model, verify that the notation is self-explanatory. A reader should be able to explain the model from the equations and variable definitions alone.

9.8 Identification Strategy: What Is Identification?

“Identification is just another term for your particular approach of estimating causal effects” (Nikolov 2013). Much empirical work boils down to a claim that A causes B, usually documented by regression. Your job is to explain how the causal effect you think you see in the data is identified.

9.9 Identification Strategy: What to Include

Nikolov (2013) recommends:

1. State your identification strategy clearly: OLS, instrumental variables, difference-in-differences, regression discontinuity, randomized control trial?
2. Assume the reader knows the basics but not how you apply the method in your context.
3. Be specific:
   • If IV: What is Y? What is X? What is your instrument? Why is it valid?
   • If RDD: What is the rule? How are participants assigned?
   • If DiD: What are the treatment and control groups? What is the identifying assumption?
4. Discuss assumptions upfront: What do you need to assume to interpret your parameters causally?
5. Write out the econometric specification: Explain each variable and parameter.

9.10 Addressing Biases

Nikolov (2013): “What are biases of a naïve estimate?” Describe what happens if you ignore key issues (omitted variables, endogeneity, selection bias) and how your strategy addresses them.

Example (adapted from Nikolov 2013):

A naïve regression of wages on education would suffer from ability bias: more able people earn more and also obtain more education. To address this, I use an instrumental variables strategy, instrumenting for education with distance to the nearest college. This instrument is valid under the assumption that distance affects education but does not directly affect wages except through education.

Rule: Before presenting results, include one paragraph that a skeptical referee would accept as a convincing case for the identification strategy.

9.11 Results: How Many Results to Report

“Less is usually more” (Nikolov 2013). A common mistake is to include every parameter estimate from every regression. This “kitchen sink approach” muddies the message. Present only those parameter estimates that speak directly to your topic.

Rule: If a table contains more than 4–6 columns, determine whether all are necessary. Move robustness checks to an appendix.

9.12 Results: Presenting Results in Tables

Table design principles (Nikolov 2013; Dudenhefer 2009, p. 53):

• Each table should have a self-contained caption.
• The caption should include the regression equation and variable names.
• No number should appear in a table that is not discussed in the text.
• Report standard errors in parentheses (not t-statistics, unless your field requires otherwise).
• Use 2–3 significant digits. Do not report whatever your software spits out.
• Use sensible units.

9.13 Results: Example Table

Table 1: OLS Estimates of the Effect of Education on Wages
Dependent variable: Log of Yearly Earnings, 1985–1995

Notes: Standard errors in parentheses. Data from the Tennessee Second Grade Ability Survey. The ability dummy equals 1 if the teacher classified the individual as able. Each regression includes yearly dummies, industry and occupation dummies, experience, experience squared, seniority, region, marital status, race, gender, and urban dummy.

9.14 Results: Describing Results in Text

Structure of a results paragraph (Nikolov 2013):

1. Topic sentence: State the main point of the table.
2. Specific findings: Walk the reader through the key numbers.
3. Interpretation: Explain what the numbers mean in economic terms.
4. Big picture: Connect the results to the overall theme of the paper.

Good example (adapted from Nikolov 2013):

Table 1 shows that including a measure of ability in the wage equation dramatically lowers the predicted effect of education on earnings. Column 1 does not include an ability measure and indicates that a year of education raises wages by 9.1 percent. Column 2 adds the ability measure; the education effect now drops to 3.1 percent. Columns 3 and 4 show that this general pattern is repeated even when state-level dummy variables are included. The estimates in Table 1 are therefore consistent with the hypothesis that the OLS estimates suffer from an upward ability bias.

Bad example:

The first column of Table 1 shows the main effect. The second column shows the ability effect. Finally, the inclusion of state dummies does not significantly affect the main estimates, though statistical precision is lost.

9.15 Statistical Significance vs. Economic Significance

Nikolov (2013): “Explain the economic magnitude of the central numbers, not just their statistical significance. Especially in large panel data sets, even the tiniest of effects is statistically significant.”

Rule: For every key coefficient, report: (1) the point estimate, (2) the standard error, (3) the statistical significance, and (4) what the number means in real-world terms (e.g., “a one-year increase in education raises wages by 3.1 percent”).

9.16 Rounding and Precision

Nikolov (2013) provides a clear rule:

• Find the first non-zero digit in the standard error.
• Report the standard error to that decimal place.
• Report the coefficient to the same decimal place, plus one additional place.

Example: 1.23456789 ± 0.203040506 → 1.2 ± 0.2

Rule: Never report more than 3 significant digits. If the raw output is 0.00456789, report 0.005.

9.17 Discussion: Policy Implications

Nikolov (2013): “You should avoid making value judgments and rely instead on economic facts and analyses.” For example, it is appropriate to say “substituting policy X for current policy Y would raise GDP by 2 percent,” but not “policy X should be substituted for policy Y.”

9.18 Discussion: Limitations

“Point out the limitations of your research… In an undergraduate term paper such limitations are expected. In general, it is better to show your instructor that you understand the limits of your method than make broad claims you do not support” (Nikolov 2013).

Rule: Include one paragraph that honestly describes the three biggest limitations of the study. This builds credibility.

9.19 Discussion: Future Research

Suggest questions or alternative approaches for further research. This shows you understand the broader context.

9.20 Robustness Checks

Move robustness checks to appendices or online appendices once you have verified they do not change the main result (Nikolov 2013). In the main text, summarize: “The results are robust to alternative specifications, including… (see Appendix A).”

Part 10: Writing a Theoretical Paper

10.1 What Is a Theoretical Paper?

In a theoretical paper, a model is extensively developed and shown to be internally logically consistent (Dudenhefer 2009, p. 30). Much like proofs in geometry, the conclusions are “proved”—the model is shown with sufficient internal logical consistency to demonstrate, for example, that an economic agent will choose one course of action over others.

Unlike empirical papers, theoretical papers can contain as much mathematics as prose on many pages. They are identified by an abundance of “propositions,” “proofs,” “theorems,” and “lemmas” and by an absence of data (Dudenhefer 2009, p. 30).

10.2 Structure of a Theoretical Paper

Theoretical papers are less standardized than empirical papers, but they typically include (Dudenhefer 2009, p. 30):

1. Introduction (brief; may include a short literature review)
2. Model (the heart of the paper; may span multiple sections)
3. Propositions, Theorems, Lemmas, and Proofs
4. Discussion / Extensions
5. Conclusion

Note: Some theoretical papers include a brief empirical test or calibration exercise, but this is not required.

10.3 Presenting the Model: Infrastructure to Superstructure

William Thomson (2001, cited in Dudenhefer 2009, p. 37) advises: “Introduce your model by moving from infrastructure to superstructure. In specifying an economy, introduce and describe each actor category separately before bringing them together.”

Example: If your model contains consumers and producers:

1. Introduce consumers first: their endowments, preferences, and information.
2. Then introduce producers: their technologies and constraints.
3. Finally, bring them together in equilibrium.

10.4 State Assumptions in Order of Plausibility

Thomson (2001) also recommends stating your most plausible and general assumptions first, moving successively to your most restrictive and least plausible.

Rule: Before presenting the model, list all assumptions. Order them from most to least plausible. Present them in that order.

10.5 Verbal Explanation Before Mathematics

Every equation should be introduced verbally. Dudenhefer (2009, p. 37): “The description should begin verbally.” Explain the economic intuition before writing the math.

Example (Krugman 1991, cited in Dudenhefer 2009, p. 37):

We consider a model of two regions. In this model there are assumed to be two kinds of production: agriculture, which is a constant-returns sector tied to the land, and manufactures, an increasing-returns sector that can be located in either region. All individuals in this economy are assumed to share a utility function of the form…

10.6 Define Variables Immediately

After presenting an equation, define every variable right away.

Example:

U = C_M^μ C_A^(1-μ)

where C_A is consumption of the agricultural good and C_M is consumption of a manufactures aggregate.

Rule: Never let an undefined symbol appear more than one line after its first use.

10.7 Notation

Dudenhefer (2009, p. 37):

• Follow standard notation in the literature.
• If you deviate from standard notation, make your notation self-explanatory.
• Use subscripts, superscripts, and Greek letters liberally.
• Use equation editors for readability.

10.8 Propositions, Theorems, and Proofs: Structure

A typical pattern (Dudenhefer 2009, p. 30):

1. State the proposition or theorem clearly.
2. Provide the proof.
3. Explain the economic intuition behind the result.

Rule: After every proof, include one paragraph explaining the intuition in plain English. Explain what the result means for economic behavior.

10.9 Clarity in Proofs

McCloskey (2000, p. 47) warns against raising “the symbolic ante”: “I am turned off when I see only formulas and symbols, and little text. It is too laborious for me to look at such pages not knowing what to concentrate on” (Ulam 1976, cited in McCloskey 2000, p. 47).

Rule: For every line of a proof, verify that a reader would understand why this step follows from the previous one without reading three pages ahead. If not, add explanatory text.

10.10 Examples of Theoretical Papers

Dudenhefer (2009, p. 30) cites several famous theoretical papers as models:

• Solow, Robert. 1956. “A Contribution to the Theory of Economic Growth.” Quarterly Journal of Economics 70 (February): 65–94.
• Akerlof, George A. 1970. “The Market for ‘Lemons’: Quality Uncertainty and the Market Mechanism.” Quarterly Journal of Economics 84 (August): 488–500.
• Krugman, Paul. 1991. “Increasing Returns and Economic Geography.” Journal of Political Economy 99 (December): 483–99.
• Yildirim, Huseyin. 2006. “Getting the Ball Rolling: Voluntary Contributions to a Large-Scale Public Project.” Journal of Public Economic Theory 8 (4): 529–48.

Rule: Model the presentation on established theoretical papers in the field. Note how they present assumptions, structure proofs, and explain intuition.

10.11 Style Considerations

Avoid Boilerplate

McCloskey (2000, p. 36): “The academic pose inspires boilerplate. Little is getting accomplished with econometric chatter copied out of the textbook, rederivations of the necessary conditions for consumer equilibrium, and repetition of hackneyed formulations of a theory.”

Be Concrete

Even in theory, concreteness helps. Instead of “two cities, A and B, trading an asset, X,” write “New York and London in 1870 both had markets for Union Pacific bonds. The question is, did the bonds sell for the same in both places?” (McCloskey 2000, p. 36).

Rule: After generating the model, replace at least three abstract terms with concrete examples.

Part 11: Designing Tables and Figures

11.1 General Principles

Tables and figures are not afterthoughts—they are part of your argument.

McCloskey (2000, p. 46): “The wretched tables and graphs in economics show how little economists care about expression. Tables and graphs are writing, and the usual rules of writing therefore apply. Bear your audience in mind. Try to be clear. Be brief.”

Key principles:

• Every table/figure must be discussed in the text (Nikolov 2013).
• Every table/figure should be self-contained: a skimming reader should understand it without reading the text.
• Less is more: include only what is necessary (McCloskey 2000, p. 46).

11.2 Tables: Parts of a Table

Dudenhefer (2009, p. 53) identifies the main parts:

• Table number: Consecutive throughout the document (Table 1, Table 2, etc.)
• Title: Brief but descriptive. Not a complete sentence.
• Column heads: Identify the information in each column
• Stub: The left-most column
• Body: The data columns
• Footnotes: Source notes, general notes, specific notes
• Rules: Horizontal lines separating parts (avoid vertical rules)

11.3 Tables: Title and Notes

Title should state the subject clearly:

• Good: “Effect of Class Size on Student Achievement: OLS Regression Results”
• Bad: “Regression Results” (too vague)

Nikolov (2013): “The caption of a regression table should have the regression equation and the name of the variables, especially the left-hand variable.”

Notes are essential. Nikolov (2013): “The notes to your table should be extensive enough so that the reader does not have to look back at the text to understand what is being presented.”

Every table should indicate:

• Source of the data (if not yours)
• Definitions of variables
• Standard errors or t-statistics (specify which)
• Sample restrictions
• Additional controls included

11.4 Tables: Rounding and Variable Names

Rounding:
Nikolov (2013): “Use the correct number of significant digits, not whatever the program spits out. 4.56783 with a standard error of 0.6789 should be 4.6 with a standard error of 0.7. Two to three significant digits are plenty.”

McCloskey (2000, p. 46): “Who wants to read 3.14159256 when 3 1/7 describes the elasticity without making the reader stop to grasp the stream of numbers?”

Variable Names:
McCloskey (2000, p. 46): “In headings of tables you should use words, not computer acronyms. A column labeled ‘LPDOM’ requires a step of translation to get to the meaning: ‘Logarithm of the Domestic Price.’”

Rule: Replace every acronym in tables with the full term or a clear abbreviation.

11.5 Tables: Minimum Information

Dudenhefer (2009, p. 53): A table should contain at least two columns and six cells of information. If you have fewer, present the information in the text instead.

11.6 Figures: When to Use

Nikolov (2013): “Good figures really make a paper come alive, and they communicate patterns in the data much better than big tables of numbers.”

Use figures for:

• Trends over time
• Distributions
• Relationships between variables (scatter plots)
• Maps or spatial patterns

11.7 Figures: Design Principles

Nikolov (2013):

• Give a self-contained caption, including a verbal definition of each symbol.
• Label the axes.
• Use sensible units.
• Don’t use dotted line types that are invisible when reproduced.
• Don’t use dashes for very volatile series.

McCloskey (2000, p. 46), following Tufte (1983, 1990):

• “Mobilize every graphical element, perhaps several times over, to show the data.”
• Use titles that state the theme: “All Conferences Should Happen in the Midwest” rather than “A Model of Transport Costs.”
• Use meaningful names for lines and points, not alphanumeric codes.

11.8 Self-Contained Captions

Nikolov (2013): “Each table should have a self-contained caption so that a skimming reader can understand the fact presented without having to go searching through the text for things like the definitions of Greek letters.”

Rule: Generate captions as if the reader will see only the caption and the table/figure, with no other text.

11.9 Equations

McCloskey (2000, p. 47): “It’s clearer and no less scientific to say ‘the regression was Quantity of Grain = 3.56 + 5.6 (Price of Grain) – 3.8 (Real Income)’ than ‘the regression was Q = 3.56 + 5.6P – 3.8Y, where Q is quantity of grain, P its price, and Y real income.’”

Paul Halmos (1973, cited in McCloskey 2000, p. 47): “The author had to code his thought in [symbols] (I deny that anybody thinks in [such] terms), and the reader has to decode.”

Rule: For every displayed equation, verify that a reader who skipped the text can still understand what it means. If not, add explanatory text or use words instead of symbols.

11.10 Common Mistakes

11.11 The “So What?” Test for Visuals

For every table and figure, ask:

1. What is the one thing I want the reader to take away from this?
2. Is that takeaway obvious at a glance?
3. Is this better shown visually than described in text?

If the answer to #3 is no, consider presenting the information in prose instead.

Part 12: Writing the Conclusion

12.1 What a Conclusion Should Do

A conclusion should:

1. Restate your main findings briefly
2. Discuss policy implications (without normative judgments)
3. Acknowledge limitations
4. Suggest directions for future research

Nikolov (2013): “Really, a conclusions section should not be necessary. If you did a good job of explaining your contribution in understandable prose in the introduction, and then documenting those claims in the body of the paper, then saying it all over again is pointless.” However, readers have come to expect a conclusion, so provide one—but keep it short.

12.2 Length

• Undergraduate term paper: About 1 single-spaced page (Nikolov 2013)
• Journal article: Typically 1–2 paragraphs

Nikolov (2013): “Conclusions should be short and sweet. Do not restate all of your findings. One statement in the abstract, one in the introduction and once more in the body of the text should be enough!”

12.3 Content: Main Findings

Restate your central result in one sentence. Do not list every coefficient or table.

Example:

This paper finds that including a measure of ability in the wage equation dramatically lowers the predicted effect of education on earnings, from 9.1 percent to 3.1 percent per year of schooling.

12.4 Content: Policy Implications

Nikolov (2013): “You should avoid making value judgments and rely instead on economic facts and analyses.” It is appropriate to say what would happen under a policy change; it is not appropriate to say what should be done.

Good: “Substituting policy X for current policy Y would raise GDP by 2 percent.”

Bad: “Policy X should be substituted for policy Y.”

Why: Measuring welfare consequences is difficult. Dollars are easy to measure, but well-being is not. Your research may not have accounted for distributional issues, legal issues, or matters of national sovereignty (Nikolov 2013).

12.5 Content: Limitations

“Point out the limitations of your research, say the relatively small number of observations you have or the simplicity of the functional form you have tested. In general, it is better to show your instructor that you understand the limits of your method than make broad claims you do not support” (Nikolov 2013).

Rule: Include one paragraph that honestly states the three biggest limitations of the study. This builds credibility with readers and referees.

12.6 Content: Future Research

Suggest questions or alternative approaches. This shows you understand the broader research agenda.

Example:

Future research could explore whether these findings hold in developing countries, where credit markets may function differently. Additionally, a dynamic model that allows for adjustment costs could provide further insights into the timing of firm responses.

12.7 What NOT to Do

Do Not Repeat the Introduction

Dudenhefer (2009, p. 51): “The conclusion is your chance to sum up your argument in a clear and concise manner, and in a way that does not simply repeat, word for word, what has been already said.”

Do Not Introduce New Results

The conclusion is not the place to present findings that should have been in the results section.

Do Not Speculate

Nikolov (2013): “Don’t speculate; the reader wants to know your facts not your opinions.”

Do Not Write a Table of Contents

McCloskey (2000, p. 37): “Another piece of boilerplate… is the table-of-contents paragraph: ‘The outline of this paper is as follows.’ Don’t, please, please, for God’s sake, don’t.”

12.8 The Conclusion as an Inverted Introduction

Dudenhefer (2009, p. 51): “Conclusions are, in a way, upside-down versions of introductions: whereas in introductions you usually build up to your thesis statement, in conclusions you usually begin with it.”

Structure:

1. Thesis / main finding (first sentence)
2. Brief recap (one sentence)
3. Implications (one paragraph)
4. Limitations (one paragraph)
5. Future research (one sentence or paragraph)

12.9 Consistency with the Introduction

Dudenhefer (2009, p. 51): “I would suggest reading your introduction and your conclusion side by side. They should be consistent with one another: the thesis or question or conclusion you state in your introduction should be the one you state in your conclusion.”

Rule: Verify that the last paragraph of the introduction and the first paragraph of the conclusion are consistent. If they do not say the same thing, revise one or both.

12.10 Final Sentence

End with a sentence that looks forward, not backward. Dudenhefer (2009, p. 51): “While looking back at the paper just presented, the conclusion should also look ahead.”

Good examples:

• “These findings suggest that future research should focus on…”
• “Understanding [X] remains an important challenge for economists.”

Bad examples:

• “In conclusion, this paper has examined…”
• “To sum up, we found that…”

Part 13: Citations and References

13.1 In-Text Citations: Two Styles

Economics uses the author-date (Harvard) citation system. There are two main styles: loud references and soft references (Nikolov 2013).

Loud Reference

Use when you wish to acknowledge the source of an idea explicitly. Cite the name of the author(s) in the body of the sentence and place the publication date in parentheses.

First mention (full name):

Sender-receiver games, introduced by Jerry Green and Nancy Stokey (1980) and Vincent P. Crawford and Joel Sobel (1982), provide the simplest stylized environment in which communication is essential.

Subsequent mentions (last name only):

These theories come in two guises: explicit dynamic theories, i.e., Canning (1992) and Nöldeke and Samuelson (1992), and static solution concepts, i.e., Blume et al. (1993) and Wärneryd (1993).

Rules:

• Use “et al.” when authors number three or more.
• Place punctuation after the parenthetical date.

13.2 Soft Reference

Use when you wish to evoke a source that substantiates a claim. Cite the name(s) and date in parentheses.

First mention (full name):

Such babbling equilibria are proper (Roger B. Myerson, 1978; Blume, 1994), and even strategic stability (Elon Kohlberg and Jean-Francois Mertens, 1986) does not rule out uninformative equilibria in general.

Subsequent mentions (last name only):

Such babbling equilibria are proper (Myerson, 1978; Blume, 1994), and even strategic stability (Kohlberg and Mertens, 1986) does not rule out uninformative equilibria in general.

Rules:

• Authors are listed in order of publication date.
• Separate sources with a semicolon.

13.3 Listing References

References are listed alphabetically by the last name of the first author (Nikolov 2013).

Formatting Author Names

Note the semicolon used to divide some, but not all, names in the three- and four-author examples.

13.4 Repeat Authors and Page Numbers

Repeat Authors:
If you cite multiple works by the same author(s), do not spell out the name(s) after the first entry. Use an underscore (___) to signify the name(s). List sources by the same author(s) in order of publication date.

Example:

Krugman, Paul R. and Venables, Anthony J. Integration and the Competitiveness of Peripheral Industry, in Christopher Bliss and Jorge Braga de Macedo, eds., Unity with diversity in the European economy. Cambridge: Cambridge University Press, 1990, pp. 56–77.

___. Globalization and the Inequality of Nations. Quarterly Journal of Economics, November 1995, 110 (4), pp. 857–80.

If two sources by the same author were published in the same year, label them 1990a and 1990b.

Page Numbers:
Page numbers are inclusive. If the first and last page numbers share an initial numeral, drop the initial numeral of the last page number: pp. 367–98 (not pp. 367–398).

13.5 Types of Sources

Journal Article

Format: Author. Title of the Article. Name of the Journal, Month Year, Issue Numeral (Issue Number), pp. XX–XX.

Examples:

• Davis, Donald R. Intra-Industry Trade: A Heckscher-Ohlin-Ricardo Approach. Journal of International Economics, November 1995, 39 (3–4), pp. 201–26.
• Kohlberg, Elon and Mertens, Jean-Francois. On the Strategic Stability of Equilibrium. Econometrica, September 1986, 54 (5), pp. 1003–37.

Book

Format: Author. The Title of the Book. City of Publication: Name of the Press, Year of Publication.

Examples:

• Rosenberg, Nathan. Perspectives on Technology. New York: Cambridge University Press, 1976.
• Helpman, Elhanan and Krugman, Paul R. Market Structure and Foreign Trade. Cambridge, MA: MIT Press, 1985.

Chapter in an Edited Volume

Format: Author. “Title of the Chapter,” in Editor(s), ed(s)., Title of the Book. City: Publisher, Year, pp. XX–XX.

13.6 Common Mistakes

13.7 Footnotes

McCloskey (2000, p. 48): “A footnote should be subordinate. That is why it is at the foot.” Footnotes should guide the reader to sources—not substitute for good arrangement. Avoid breaking sentences with footnotes (McCloskey 2000, p. 48).

Nikolov (2013): “Put details and digressions in footnotes.”

Rule: Before adding a footnote, determine whether it belongs in the main text. If yes, move it. If no, keep it brief.

13.8 Citation Management

Rules:

1. Ensure every citation is accurate and complete. Do not rely on secondary sources for primary claims.
2. Follow one style guide (e.g., AER style) consistently throughout the document.
3. Verify every citation against the original source before finalizing.

Part 14: Final Quality Checklist

Final Quality Checklist

Before delivering the paper, run through this checklist:

• Does the paper have one clear, central contribution stated in the abstract and introduction?
• Is the paper organized in “newspaper style” (most important first)?
• Are all sentences clear on first reading?
• Are all subjects short and all verbs active?
• Is every paragraph a complete discussion of one topic?
• Are all numbers rounded appropriately?
• Are all tables and figures self-contained?
• Are all sources cited using author-date format?
• Have you cut at least 20% of the words from the first draft?
• Is the tone reasonable and modest, not pompous?

Key Takeaways

1. Clarity is king — Every sentence must be clear on first reading
2. Be concise — Cut 20% of words from your first draft
3. Know your audience — Write for busy, impatient readers who skim
4. Use active verbs — Make the doer the subject and the action the verb
5. Show, don’t just tell — Use concrete examples, not abstract jargon
6. Structure matters — Organize like a newspaper (inverted pyramid)
7. Quality over quantity — Fewer, better results beat a kitchen sink
8. Be generous — Criticize previous work gently
9. Look forward — End conclusions with forward-looking sentences
10. Check twice — Run through the quality checklist before submitting

Thank You

Based on:

• Dudenhefer, Paul. 2009. A Guide to Writing in Economics. Duke University.
• McCloskey, Deirdre N. 2000. Economical Writing. 2nd ed. Waveland Press.
• Nikolov, Plamen. 2013. “Writing Tips For Economics Research Papers.” Harvard University.
• Wallwork, Adrian. 2011. English for Writing Research Papers. Springer.