As someone who reads a lot of SoP's, here are the questions I want an applicant to answer, more or less in order:

• What is your purpose? What are your career aspirations?  What type of research do you want to conduct?  What topics and methods are you intending to pursue?
• What led you to your purpose? Why are you applying to this field?  This is a big issue in accounting, as many applicants could well have applied to programs in less lucrative and/or more competitive fields, like economics, psychology, finance or statistics. You need to convince the faculty that you are pursuing your purpose because you are truly interested in the subject material in your chosen field.   Research and teaching are simply too much work for anyone to succeed if they aren’t inherently interested.
• Why will you succeed? The faculty already have your transcript and a CV that lists your accomplishments.  This is your chance to pull those facts together into a story that answers the questions that faculty are going to be most concerned about:  Do you know enough about accounting (or whatever field you have chosen) to ask interesting questions?  Are you smart enough?  Can you write?  Are you sufficiently well-trained in the fundamentals (math, statistics, psychology, depending on your purpose)?  Will you work hard enough?  Do you have the emotional makeup to learn and improve from constructive criticism (which you will receive in abundance throughout your career)?   But you can’t just write “I am smart and hard working.”  Follow the classic writer’s advice:  show me, don’t tell me.  You’ll be much more persuasive writing “As an undergraduate I regularly took 18-credit terms, still averaging a 3.6 GPA in statistics and calculus” or “During my 3 years at East Coast Investment Bank I learned how misleading accounting reports can be.”
• Do you know what you are getting into? Many highly-qualified applicants are admitted but fail to thrive because they don’t actually want to do the job they applied for!  Being a PhD student or faculty member involves spending a lot of time alone with your computer, reading, writing, trying to find something new to tell people who already know a great deal.  You’ll have to put up with constant criticism–and learn from it.  Even though you are in accounting, you will spend most of your PhD years studying foundational material that is far removed from the applications that interest you.  When you finally conduct research on a topic of your choosing, it will be far narrower and more distant from real-world issues than you are probably expecting.   You might address this point in a specific paragraph, but an alternative is to weave in evidence throughout the entire SoP that demonstrates that you know what you are getting into, and you like it.
• Have you ever conducted research? If you have, you can use that fact to address many of the above questions.  Perhaps a research project with a professor led to your interest in accounting.  Perhaps the paper provides evidence of hard work and math skills.  Perhaps it shows that you know what you are getting into.  There is no better way to show that you can succeed at conducting research than to have succeeded at conducting research! You probably need a separate paragraph to describe your research topic, method and conclusions.  But place that discussion in the context of one of the above bullet points.
• What are your weaknesses and how will you address them? Almost no one has a strong answer to every question in the previous bullet points.  If you lack a strong math background or work experience, faculty will think more of you if you acknowledge that fact (after all, they already know from your transcript) and propose a plan for dealing with it.  This shows both self-awareness and an understanding of what the job requires.
• Why us? You could pursue your purpose at any number of institutions.  Why are you applying to my school?  I strongly recommend reading articles posted on SSRN written by faculty at every school you apply to.  Here is your opportunity to explain which faculty you are hoping to work with, and why those people and the program are attractive to you.
• Why you? Doctoral program admission rates are extremely low in accounting and other management fields (under 5%, usually just one or two people per field per year).  There are probably other applicants who are similarly qualified.  Why should the school admit you?  This is your opportunity to explain why you are special.  Everyone has some experience or training that will make them more memorable and suggest qualifications above and beyond the usual qualifications.  Use yours to stand out from the crowd, and as a means of drawing your SoP to the only possible conclusion–that the school should admit YOU.

A final remark on writing style.  Your SoP is also a way to convey that you understand how academics write.  Here are some rules successful doctoral students learn quickly:

• Don’t tell us a mystery story. If you are writing fiction, keep me in suspense.  But if you are writing an academic paper–or a Statement of Purpose.  I have to read about 100 SoPs in a typical admissions cycle.  Don’t make me wade through any more words than necessary before telling me what I need to know.
• Don’t tell us “what I did for summer vacation.” This is particularly important for academic research.  We want to know the conclusion, how you support it, what you found and why it matters.  We don’t want to know the fifteen different ways you tried to solve the problem before you found one that works.  Your SoP might deviate from this rule a little bit, because we want to know why you ended up applying to our program.  But even for that insight, what we really want to know is why right now, after all you have done and learned, you want to apply to our program.
• Don’t overclaim. This cardinal rule of academic writing has three components.  First, never overstate the facts.  Don’t say that you were the best student in your class if you weren’t.  Any hint of overclaiming facts is going to raise questions about your academic integrity, and might be grounds for immediate disqualification.  Second, don’t overstate the implications of the facts.  Being first in your class doesn’t mean you are sure to succeed, so don’t say it does. Third, don’t overstate the value of the facts.  I wouldn’t recommend writing that being in first in your class is “extremely impressive”.  Many academic writers will advise you never to say that a certain fact is interesting, important, impressive, astounding, crucial, or the like.  Instead, just say that the fact implies a certain outcome or conclusion, and let the reader judge whether that is interesting,  important and impressive.  After all, you wouldn’t have even mentioned it if you didn’t think it was interesting, important and impressive.  Saying so explicitly just makes the academic reader want to take issue with you.  (This is how we roll.)  Just lay out the facts and their direct implications in a way that allows no conclusion other than “wow, that is interesting, important and impressive!”

Good luck!]]>

As I've been considering this, I've thought about the different ways in which standard-relevant research can be conducted. I've realized there are implementation studies and there are theory-based studies. I'll demonstrate using papers we discussed in a PhD seminar this past spring. Obviously each paper contains elements of each category, but I focus on them in a way to emphasize my point.

Implementation Studies: Some papers examine specific proposals for standards. Research questions include, “Does the new proposed standard on XXXX improve decision-making or affect investor reactions?” Wainberg et al. (2011) have a paper that looks at recent changes to PCAOB inspection reports. They compare user interpretation of the old format of PCAOB reports, the new format of PCAOB reports, and three hypothetical PCAOB reports. The results of the paper are directly applicable to the specific standard setting project at hand: which format type is best for PCAOB inspection reports.

The outcomes of this study are of immediate interest to current standard-setting projects. However, papers that focus strictly on implementation may put the authors in a bit of a time-crunch: will the study still be relevant if there are delays (what if they miss the comment window)? Will the study still be interesting if the standard setters change the proposed formats? Will the study still be publishable if the proposed standard is dropped?

Theory-based Studies: Focusing in on a specific issue can be interesting, but it can also be risky. One way to reduce this risk is to map testable elements into broader constructs. Many papers examine a specific issue and then try to generalize the results. The Wainberg et al. (2011) paper does just that. It looks at a very specific issue, but then discusses how the results are of a more general interest. I think the paper would be extremely interesting, even if the PCAOB hadn’t been considering alternative report formats.

Other papers try to ask what general principles may help in understanding a proposed standard, and then design a study to examine those principles, often in a context other than the proposed standard. For example, a paper by Bowlin, Hales, and Kachelmeier (RAST 2009), uses a simple economic game to better understand how a manager’s prior experience as an auditor may affect strategic reporting decisions. This paper has interesting implications for the recent PCAOB discussion on auditor independence. Similarly, a paper by Holderness and Hunton (2011) considers the relationship between earnings management and forecast disaggregation, but at a level that does not tie the study to a specific proposed standard. Nonetheless, the paper is extremely relevant in considering management forecast discretion.

As I’ve considered various standard setting projects, I sometimes think, “I wouldn’t have time to test that before comments are due.” However, as I have thought about theory-based studies, I have realized that the window expands. Instead of being constrained to consider a specific proposal, I can look at a standard setting agenda (for example the PCAOB agenda that was released today), and think about theory that may be of interest to standard-setters in a couple of years. This gives me time to get a good project going, one that will be of interest to standard setters while still being of general interest to researchers, even if any specific proposed standard is dropped.]]>

How many of you work for a university that only values “A-hits”?  To the non-academics, this mean that a publication in one of 3 or so top journals counts as meeting the research expectations placed on the faculty member.  Publication(s) in other journals do not count.  Recently, my institution elevated the value of A-hits, but other publications still count some.  My impression from talking to colleagues is that many other schools are valuing A-hits only.

I do NOT want this post to degrade into journal/editor bashing.  The top journals have earned a reputation amongst us academics.  Rather the issue to me is the long-term ramifications for accounting research from this move by deans and provosts to (a) judge research quality purely by journal quality, and (b) to assign extremely low weights to publications in good journals that do not happen to be on the school’s A list.  The issue came up during the PhD Coordinators’ breakfast at AAA.  Several folks spoke eloquently on how this could really curtail the scope of inquiry in accounting research and make it almost impossible for new and novel areas of research to develop.  Again, I am not saying a school should value every publication equally; the concern is putting zero value on something that later proves to be a key innovation in our discipline, with the zero value arising simply because the research was not quite what the A journals wanted to publish.

Now, the tie back to policy research.  Holly Skaife mentioned in her reply to my post about the IASB “As I stated in the session, the information needs of the IASB and IFRS Foundation include institutional details or basic descriptive evidence that is not always sufficient for publication in our academic journals (e.g., what are the pension laws in countries that adopt IFRS and do those laws mandate pension funding requirements).”  Actually, some journals are willing to publish important descriptive evidence about practice.  Accounting Horizons has valued that sort of paper.  But a publication in Horizons might not be counted by the faculty member’s dean.

I want to encourage folks who want to work on policy related research to keep working, even if that research might be destined for less than an A journal.  To back up my encouragement, I want to replay an exchange that occurred in a panel session at the AAA.

Audience member:  we academics publish interesting work that is relevant for the standard setter, but because it is not in the A journal, the standard setter does not pay attention to it.

Standard setter: The FASB has a subscription to a host of journals.  Many of the Board members and staff regularly receive the table of contents of each new issue, and our library furnishes us copies of articles that are relevant to our projects as needed.  This is done for 10 or 12 journals, not just the top 3.

My point is this, as a researcher, is it more important to have perceived impact (something your dean counts) or real impact (something policy makers read)?  I know you have to pay attention to the former, especially for untenured faculty.  But the latter seems a lot more important to me.]]>

My first thought when I saw this title was that the more readable the financial reports and disclosures, the more analysts would want to follow the firm. But on second thought (and as the paper’s results bear out), I wondered whether analysts choose to follow firms whose financial reports are less readable because that is where they can add value. It turns out that, even though analysts take a longer time to revise their forecasts for firms with poor readability (suggesting that low readability is tough on analysts too, not just your average investors), there are more analysts following such firms than firms with high readability. Apparently, demand for interpretive services is sufficient to pay for more analysts when financial statement readability is relatively low.

After reading this paper, I wondered what a financial reporting standard setter would conclude. If the readability of financial reports is somehow the result of standard setting, perhaps the FASB, the IASB, and/or the SEC should focus more on making disclosures more readable. But then I had a more sinister thought occur to me. When standard setters ask for input from sophisticated users of financial reports (such as analysts), should they take the input with a grain of salt?  If analysts have an incentive to seek out companies or industries where they can add value by interpreting the low-readability reports, perhaps analysts have little incentive to seek for better disclosures (let alone readability). If you made lots of money helping people interpret financial reports because those reports are hard to understand, would you want standard setters and regulators to make financial reports more readable?

Of course, this paper is not specifically about standards and regulations that cause low readability, nor is it about standard setters reliance on input from analysts. So you should take my comments here with a grain of salt too. But if readability of disclosures somehow varies across different disclosure topics (for example, if pension disclosures tend to have lower Fog index scores than stock compensation disclosures), and standard setters are wondering what sophisticated investors think about a particular disclosure, they may want to think twice about the input they get from analysts.

By the way, I should add this caveat in defense of analysts—every analyst I ever met with in my work at the FASB seemed earnest in their desire to improve financial reporting. So, I do not mean to call into question the integrity of analysts. I am only drawing attention to the incentives some analysts may face when their bread and butter comes from interpreting hard to read financial reports.

One final thought: I wonder if the readability of my posts is associated with people’s willingness to comment on my posts. Hmmmm…

This was the worst with review papers. The point of review papers is to establish a coherent framework and show how literature fits together, but I often walked away thinking about the 100 individual citations and missing the framework altogether.

I think that my ability to read papers has improved a little over the last year (although it still has a long ways to go). So, with the hope of enlightening other similarly-distressed PhD students and preventing distress by future PhD students, I thought I’d share what I’ve learned.

1. Don’t read the paper in the order it is presented. Read the abstract first, then the conclusion. At this point, ask yourself what the author’s main objective is.
2. Pay attention to headings. I’m so used to novels that it’s in my nature to skip chapter titles. Don’t do it! In fact, if after reading the conclusion you are still unsure about the goal of the paper, skim the paper reading only the headings (and maybe hypotheses, graphs, and tables).
3. Try to summarize the entire paper in a single 3×5 notecard. This will be great when studying for comprehensive exams, but will also be extremely beneficial now in forcing you to identify the big picture(s) of the paper.
4. Theory is way more important than statistics. I quote from an e-mail conversation I had with Rob Bloomfield,

Me: “I’m afraid I’m much better at math than I am at thinking about how to write a paper”

Rob: “Sadly, the math (or data analysis) is the easy part!”

Don’t spend a lot of time thinking about the econometrics of papers: what they could have/should have done differently. The truth is, you don’t get many publications by replicating someone’s paper using a more advanced econometric technique. Nor do many workshops discuss statistics.

I would love to hear advice from others who have passed through (and hopefully overcome) this same struggle with understanding papers.

I am just starting to think about some of the questions I would want to ask these awesome panelists (all of them either current or former members of the FASB or IASB), and I would love to invite others to chime in with questions they would like to ask. Have you ever wondered how much standard setters actually rely on research when formulating new standards? Have you ever wondered what research standard setters have found most helpful in their work? Have you ever wondered what research has been least helpful? What would you love to know from these panelists about what they wish researchers knew (better)?

I invite you to post your questions here, and I will do my best to include them among the questions asked during the panel session. If you would like me to ask your question anonymously, you can email me directly and I will not mention your name when asking the question. I look forward to reading your questions and seeing you at the AAA meetings.

 To mitigate the impact of amortizing the large unrealized losses to income, companies have taken at least one of at least two strategies.  First, some companies have adopted a new proforma bottom line called EBITDAP or earnings before interest, taxes, depreciation and pension income.  Examples include GenCorp (2010), YRC Worldwide (2009), NCR (2010), and J.C. Penney (2010). Furthermore, some hedge fund managers use Price-to-EBITDAP ratios for purposes of pricing and comparing companies (e.g., see Street Insider, 12/30/2009).

 Second, some U.S. companies (e.g., AT&T and Verizon) changed accounting methods and began using mark to market accounting for pension assets, taking the changes in market value through net income (i.e., eliminating the corridor approach) (e.g., see Pensions and Investments, 2/21/2011). This second approach results in retroactively applying the pension losses to the year 2008, as opposed to gradually amortizing them to pension expense over many years. This approach results in recognition of large gains as the Dow has gradually recovered from the low of 7,000 to almost 13,000 today.

 If the sample is large enough, it would be interesting to analyze the factors leading some U.S. companies to change to mark-to-market accounting for pension assets, while others continue using the corridor approach. It would be interesting also to investigate whether the market attaches higher values to companies with large pension gains (in 2009 and 2010) as a result of the switch to mark-to-market accounting, as compared to companies reporting net losses due to continued use of the corridor approach.  Any thoughts?

 Essentially, the new IASB rules eliminate the smoothing mechanisms currently present in U.S. GAAP, including the elimination of the need to estimate the expected return on plan assets.  As a result, Zion says, “companies won’t have to worry about the earnings impact of their pension asset allocation decisions and can instead focus more on the underlying economics and risks. That could result in an increased emphasis on liability-driven-investing, which could change asset allocations” …  sounds to me like a good research question for academic accounting researchers to investigate.

 The article provides a good summary of what’s new about pension accounting as prescribed by the IASB amendments to IFRS 19.  Among other things, the article explains how much of the change in the value of the net pension asset or liability will hit net income and how much will hit OCI.  You can read the full article by clicking here.