Research for Accounting Policy - PowerPoint PPT Presentation

Loading...

PPT – Research for Accounting Policy PowerPoint presentation | free to download - id: 76a2ed-ZDBjZ



Loading


The Adobe Flash plugin is needed to view this content

Get the plugin now

View by Category
About This Presentation
Title:

Research for Accounting Policy

Description:

Research for Accounting Policy Shyam Sunder Yale School of Management October 19, 2010 – PowerPoint PPT presentation

Number of Views:26
Avg rating:3.0/5.0
Slides: 57
Provided by: ss454
Category:

less

Write a Comment
User Comments (0)
Transcript and Presenter's Notes

Title: Research for Accounting Policy


1
Research for Accounting Policy
  • Shyam Sunder
  • Yale School of Management
  • October 19, 2010

2
An Overview
  • Social sciences as instruments of social
    engineering (policy)
  • Scope and levels of accounting policy
  • Maintained assumptions of financial reporting
    policy
  • Instruments of social sciences what works, what
    doesnt? Why and why not?
  • Way forward accounting policy beyond social
    sciences?

3
We Could Talk About
  • Social science and social engineering
  • Robustness of findings to their discovery
  • Field data Causal direction, correlation of
    hypotheticals
  • Experiments and bench testing
  • Scope and levels of accounting and financial
    reporting policy
  • Criteria
  • Assumptions
  • Problems of efficient standards and their
    alternatives
  • Command and control and alternatives
  • Social science as instrument of policy
  • Accounting policy beyond social sciences
  • Challenges for policy research, historical
    experience
  • Closed or open systems?

4
Social sciences as instruments of social
engineering
  • Social science as the dominant model of business
    research since mid-20th century
  • Although engineering originated well before the
    origins of natural sciences, advances in natural
    sciences have enabled us to engineer great many
    artifacts we consider indispensable today
  • Accounting, too, originated well before policy
    research but it does not seem unreasonable to
    think that we can also take advantage of learning
    from social sciences to make better social policy
    (maintained assumption of this session)

5
Limits of Parallels
  • This parallel can take us only so far
  • Natural sciences search for, and identify, laws
    of nature valid across time and space
  • Validity, replicability, and predictability of
    these laws confers prestige on sciences
  • But there is another form of scholarship on our
    campuses--humanities that regard the behavior of
    sentient beings as infinitely variable each of
    character in Iliad, Macbeth, or Ramayana is
    unique
  • They pursue eternal truths, but admit no laws

6
What are social sciences?
  • Social recognizes that the subject of study are
    sentient beings (with free will that humanists
    recognize), not marbles or atoms without will
    (that scientists study)
  • Science seeks the prestige associated with the
    search for eternal laws
  • Neither sciences nor humanities allow much room
    for laws of human behavior we seek in social
    sciences
  • Free will and laws of behavior do not sit well
    together we want but cant have it both ways

7
Laws of Social Sciences
  • To serve as a basis for social policy, the laws
    of social sciences must have stability (be robust
    to their own discovery)
  • Since humans learn and adapt, social science
    findings can alter behavior in ways that tend to
    invalidate the findings
  • Findings which are robust to such adaptation can
    be called laws of social sciences, and may
    serve as the basis for social policy

8
How Robust Are Our Findings?
  • Independent of the method of research we use,
    robustness of findings (to their own discovery)
    is a pre-requisite for their use as basis for
    policy
  • Like unclaimed dollar bills on side-walk, many
    findings (e.g., small firm and Monday effects)
    disappear upon being reported
  • There are other findings (e.g., determination of
    price by intersection of demand and supply) are
    robust in this sense (not merely statistically)
  • So the first pre-requisite for usefulness of any
    research findings for policy is this stability
  • Most such laws are properties of institutions,
    not behavior 1

9
Causal Link
  • Policy makers want to know if the manipulation of
    the policy variable under consideration has a
    directional (causal) link to the desired
    objective. Correlation does nothing for them.
  • Yet, the problem of establishing a causal link
    between two variables on the basis of field data
    remains largely unsolved due to endogeneity
  • Experimental methods have been presented as an
    alternative to address this problem, but they,
    too, have limitations of their own
  • Consider both approaches briefly

10
Problems of Inference from Field Data Causal
Direction
  • Labeling of correlation as cause is more of a
    rule than an exception in accounting research
    journals
  • When correctly labeled as correlation, what can
    the policy maker do with the finding?
  • To claim that the finding is consistent with
    Hypothesis X fails to point out that
  • It is also consistent with innumerable other
    hypotheses not mentioned in the report, and
  • No hypothesis has been rejected (violating the
    essence of Fisher-Neyman-Pearson framework)

11
  • It is important to note the philosophical
    difference between accepting the null hypothesis
    and simply failing to reject it. The "fail to
    reject" terminology highlights the fact that the
    null hypothesis is assumed to be true from the
    start of the test if there is a lack of evidence
    against it, it simply continues to be assumed
    true. The phrase "accept the null hypothesis" may
    suggest it has been proved simply because it has
    not been disproved, a logical fallacy known as
    the argument from ignorance. Unless a test with
    particularly high power is used, the idea of
    "accepting" the null hypothesis may be dangerous.

12
  • When a researcher writes the qualified statement
    "we found no statistically significant
    difference," which is then misquoted by others as
    "they found that there was no difference."
    Actually, statistics cannot be used to prove that
    there is exactly zero difference between two
    populations. Failing to find evidence that there
    is a difference does not constitute evidence that
    there is no difference. This principle is
    sometimes described by the maxim "Absence of
    evidence is not evidence of absence."

13
  • Attempts to educate researchers on how to avoid
    pitfalls of using statistical significance have
    had little success. In the papers "Significance
    Tests Harm Progress in Forecasting," and
    "Statistical Significance Tests are Unnecessary
    Even When Properly Done,
  • Armstrong makes the case that even when done
    properly, statistical significance tests are of
    no value. A number of attempts failed to find
    empirical evidence supporting the use of
    significance tests.
  • Tests of statistical significance are harmful to
    the development of scientific knowledge because
    they distract researchers from the use of proper
    methods.
  • Armstrong suggests authors should avoid tests of
    statistical significance instead, they should
    report on effect sizes, confidence intervals,
    replications/extensions, and meta-analyses.
  • J. Scott Armstrong

14
  • Some statisticians have commented that pure
    "significance testing" has what is actually a
    rather strange goal of detecting the existence of
    a "real" difference between two populations. In
    practice a difference can almost always be found
    given a large enough sample. The typically more
    relevant goal of science is a determination of
    causal effect size. The amount and nature of the
    difference, in other words, is what should be
    studied.
  • Hypothesis testing is controversial when the
    alternative hypothesis is suspected to be true at
    the outset of the experiment, making the null
    hypothesis the reverse of what the experimenter
    actually believes it is put forward as a straw
    man only to allow the data to contradict it. Many
    statisticians have pointed out that rejecting the
    null hypothesis says nothing or very little about
    the likelihood that the null is true.

15
Problems of Inference from Field Data
Hypotheticals and Efficiency
  • Our attempt to produce policy-relevant research
    often take the following form
  • Info. Sys. 1 ? Price system 1
  • Info. Sys. 2 ? Price system 2
  • Suppose information system 1 and price function 1
    are the status quo and the policy maker wants to
    know the consequences of changing the information
    system from 1 to 2

16
Problems of Inference from Field Data
Hypotheticals and Efficiency
  • We can gather data on accounting numbers and
    prices under status quo and estimate their
    statistical relationship, R(1)
  • In many situations, we can also calculate (or
    reasonably estimate) what the accounting numbers
    would have been under the policy alternative (the
    hypothetical)
  • If we could observe prices that would be
    generated under the policy alternative, we could
    also estimate the statistical relationship R(2)
  • Does a comparison of R(1) and R(2) help the
    policy makers?
  • If stronger R(2) implies preference for the
    policy alternative, it is trivially simple to
    push R(2) to the upper limit by simply using
    prices for accounting

17
Problems of Inference from Field Data
Hypotheticals and Efficiency
  • Of course, we are rarely so lucky as to be able
    to observe P(2)
  • An oft-used practice is to estimate the
    statistical relationship between the hypothetical
    I(2) and actually observed P(1), and then compare
    this R(2) with R(1) and suggest that a stronger
    R(2) implies the alternative to be preferred
    policy
  • Info. Sys. 1 ? Price system 1
  • Info. Sys. 2 ? Price system 1

18
Logic of Inference
Reporting System 1
Price System 1
Reporting System 2
Price System 2
19
Logic of Inference
Reporting System 1
Price System 1
Reporting System 2
Price System 2
20
Logic of Inference and Policy
  • This type of inference from field data does not
    help the policy makers
  • We like them to give us a nod to acknowledge our
    work, and perhaps even support it
  • But the logical foundations of such inference,
    and its implications for policy remain to be
    worked out

21
What about Lab Experiments?
  • Lab methods allow us to address the causality
    problem with greater confidence
  • But they also raise new challenges for use of
    findings in making of accounting policy
  • Accounting is highly institutionalized (complex
    interactions, expectations), like engineering
  • Experimental methods were developed for social
    sciences where a single simple example can
    support a general proposition about existence, or
    otherwise
  • Simpler experiments suffice for basic disciplines
    such as economics, psychology and physics, but
    not for accounting policy or bridge design
  • Time scale problem choosing points vs. functions

22
Bench Testing of Policy
  • Bench testing of an accounting policy alternative
    calls for far greater complexity in design of the
    lab experiment than in case of testing economic
    theory
  • More complex decisions, larger choice space, need
    more time (order of magnitude)
  • Compare choice of a point on a function with
    choice of a function

23
Scope and levels of accounting policy alphabet
soup
  • Financial reporting FASB, IASB, SEC, etc.
  • Audit PCAOB, ASB, IAuSB, AICPA
  • Internal Audit IIA
  • Cost accounting CASB?
  • Government GASB, MFOA, OMB, GAO, Congressional
    committees
  • Education AACSB, NASBA, IAESB, AICPA, CFA, IMA

24
Macro Policy in Accounting
  • Taxation and collection
  • National income
  • Monetary policy
  • Banking regulation
  • Securities regulation
  • Employment, wages, retirement, unions
  • Wealth generation
  • Wealth distribution
  • Government budgets and financial management
  • Government accounting (control of deficits)
  • Cost of legislative proposals

25
Scope and levels of financial reporting policy
  • Financial reporting policy is made at many
    levels, with higher levels having broader scope
  • Most common level is the choice of specific
    accounting methods and disclosures
  • At the other end of the spectrum are
    institutional and structural decisions usually
    made by Congress
  • Most research is concentrated at the micro end of
    the policy spectrum, probably because the high
    frequency of such decisions, and ex ante
    availability of data sets from comparable
    contexts
  • At higher institutional levels of policy making,
    ex ante data becomes more scarce, leaving us with
    ex post analyses that come too late to serve as
    inputs to decisions on policy alternatives

26
Policy Criteria
  • Research input for policy calls for identifying
    the policy criteria and including them in
    research design
  • While accounting policy makers proclaim
    allegiance to qualitative criteria such as
    representativeness, timeliness, neutrality,
    decision usefulness, etc., few of them find their
    way into our research designs
  • Research on representativeness and neutrality,
    for example, would require comparison of
    accounting data with their corresponding
    principals which are difficult to observe
  • Research on decision usefulness is subsumed into
    correlation with stock prices, setting aside
    questions about market efficiency and the
    interests of non-equity stakeholders

27
Assumptions of financial reporting policy
  • Subjecting them to investigation, instead of
    assuming that they hold

28
1. Universal Standards
  • Universal standards of financial reporting
    applied across time, economies, industries and
    corporate size and organizational forms best
    serve the constituent interests
  • Standardization does save costs and effort,
    (coordination vs. efficiency benefits electrical
    plugs, clothing, cars, street grids, commercial
    codes, cell phones, software)
  • When does standardization become
    counterproductive
  • How do we know where to stop?
  • Rhetoric of universal accounting standards using
    analogy of weights and measures and universal
    language

29
2. The Static Ideal
  • There exists a set of financial reporting
    standards that, once discovered and implemented,
    will induce corporations and their auditors to
    prepare the best attainable financial reports
  • Dynamics of the game between financial reporting
    and financial engineering makes any such a static
    ideal all but impossible
  • Consider leases, derivatives, design of
    transactions

30
3. People or Structure
  • If we select knowledgeable, experienced,
    self-less, public-spirited, and wise individuals
    to constitute bodies that devise accounting
    standards through deliberation and due process,
    we can improve financial reporting
  • Individuals stand where they sit
  • Much emphasis on the quality of individuals, too
    little attention to the structure of game they
    are asked to play

31
4. Design, not Evolution
  • It is possible to construct or discover better
    financial reporting standards through
    deliberation in properly organized corporate
    entities (such as the IASB, the FASB, etc.).
  • Assumes that such bodies can know the
    consequences of their actions
  • History does not support the proposition
  • Balancing Cartesian design vs. Darwinian
    evolution
  • Hayeks spontaneous emergence

32
5. Specialization in Setting Standards
  • Specialist standard setting bodies, standing
    ready to address new problems, inquiries and
    requests for clarifications help improve
    financial reporting
  • Their existence encourages a new clarification
    game targeted at them
  • They must keep a full agenda (performance)
  • Revenue and budget pressures
  • Over time, their output must accumulate to a
    thick rule book

33
6. What is Good and Bad?
  • Standard setters can tell which standards are
    better and why.
  • Little evidence that they know, or can know
  • Cost-of-capital is the result of complex
    interactions among many factors (including
    accounting)
  • To what extent can we sort these influences by ex
    ante analysis and research?

34
7. Standards Monopolies
  • Granting monopoly power in a given jurisdiction
    to standards written by a given body can help
    improve corporate financial reporting
  • Informational disadvantage of a monopoly
  • No opportunity for experimentation
  • No opportunity to learn from the experience of
    alternatives
  • No pressure to do better, or to correct errors

35
8. Competition and Race to the Bottom
  • A regime that encourages reporting entities to
    choose among the standards written by competing
    organizations (and paying them a royalty for the
    privilege) induces a race to the bottom to
    devise less demanding standards
  • Counter examples (Stock exchanges, bond rating
    services, appliance standards, college
    accreditation, bank regulation, corporate
    charters across U.S. states, etc.)
  • Research on tendency of race to the bottom (or
    top)?

36
9. Force and Effectiveness
  • Increase in the power of enforcement behind
    authoritative standards improves compliance and
    quality of financial reporting
  • Increased enforcement also increases resources
    devoted to evasion
  • Do draconian punishments induce better behavior
  • Comparison of evidence from domains of crime,
    alcohol and drug abuse

37
10. Effectiveness of Statutory vs. Common Law
Approaches
  • The quasi-statutory approach to setting
    accounting standards dominates a common law
    approach to financial reporting
  • Evidence?
  • Constitution (U.K., U.S., Europe)

38
11. Written Standards Dominate Social Norms
  • Written standards backed by power of enforcement
    work better than unwritten social norms backed
    only by internal and external informal sanctions
  • Social norms govern great parts of our lives
    including many aspects of law
  • Insider trading
  • Guilty beyond reasonable doubt
  • Private commercial codes (cotton, diamond trades)

39
12. Who defends the middle ground?
  • The ideal accounting regime would consist of all
    written standards or all social norms
  • Easier to make the extreme cases for standards or
    norms alone
  • Difficulty of defending the middle ground where
    both may co-exist, as they do in many other
    aspects of life
  • Most of our models tend to be linear and
    monotonic in decision variables

40
13. What is New? Historical Analysis
  • Did financial reporting and governance problems
    originate in the 20th century
  • History tells us otherwise
  • E.g., governance problems of the East India
    Company
  • Clive, Hastings, and the Companys Court of
    Directors
  • Evolution of internal directives, financial
    reporting and auditing

41
14. Financial Reporting is Getting Better
  • Has eighty years of standardization of financial
    reports (in U.S.) helped improve the quality of
    financial reporting?
  • Evidence?
  • Is a thicker (or thinner) rule book indication of
    better financial reporting?
  • Perfect correlation between accounting and stock
    returns?
  • How do we judge if our financial reports are
    getting better?

42
15. Fewer Alternatives, Better Reports
  • Fewer the alternative treatments the reporting
    entities are allowed to choose from, the better
    the quality of financial reporting
  • Fewer alternatives also tie the hands of the
    management of well-run companies who may wish to
    signal their confidence, competence and prospects
    by choosing reporting practices others find
    difficult to emulate
  • Spence on signaling

43
16. Auditors Bargaining Power
  • Do well-specified standards enhance the
    bargaining power of the auditor vis-à-vis the
    client? Reduce their legal liability?
  • Standards also encourage clients to demand show
    me the rule
  • Reduced reliance on judgment
  • More detailed the standards, greater the part of
    accountants work that can be replaced by a
    computer, and lower the value of the service

44
17. Accounting Auditing Games
  • Written standards constrain the tendency of
    managers, auditors and investment bankers to play
    accounting and auditing games
  • On the contrary, they encourage and facilitate
    game-playing by reducing uncertainty about what
    is, and is not, acceptable
  • 3 percent SPEs gt Enron
  • Does codification of GAAP serve as a roadmap for
    evasion to guide the financial engineers?

45
18. Individual responsibility
  • Written financial reporting standards strengthen
    the individual responsibility of managers,
    auditors, and investment bankers for fair
    representation
  • On the contrary, they may undermine individual
    responsibility for fair representation and the
    big picture by shifting attention to meeting the
    letter, not spirit, of the specific provisions
    and their wording
  • John C. Burton (1975) on true and fair override

46
19. Education
  • Written standards make it easier to educate
    better accountants and attract talent to the
    profession
  • Written standards may also degrade the class room
    from reasoning and intellectual debate to rote
    memorization, reinforce street image of
    accounting as boring and mechanical
  • They make it less attractive to young talent

47
The Problem of Setting Efficient Standards
Calling for Research
  • Criteria
  • Generation of alternatives
  • Evaluation of alternatives
  • Complex interactions among accounting, capital
    and labor markets financial engineering
  • Facilitation of evolution of accounting norms
  • Balancing statutory and common law
  • Balancing adjustment speed and errors of policy
  • Extent of standards, and consequences for
    personal responsibility
  • Command-and-control (nanny knows all) or bottom
    up (laissez faire) or some combination of the
    two? What combination?

48
Command Control Perspective
  • To develop accounting standards
  • A single set (monopoly?)
  • Of high quality (what does that mean?)
  • Understandable (to who?)
  • Enforceable (stick, not social norms)
  • Global (no clientele or diversity)
  • Are we willing to explore alternative mind sets
    about financial reporting through research?

49
Instruments of social sciences
  • Analysis of data gathered from the field
  • Analysis of controlled experiments in the lab or
    the field
  • Abstract mathematical analysis
  • Historical analysis
  • Introspection
  • If accounting research is to contribute to
    policy, we shall have to use all the tools at our
    disposal as and when necessary
  • Prior commitment to one or the other tool set is
    likely to be self-defeating

50
Way forward accounting policy beyond social
sciences?
  • Accounting policy includes considerations that go
    beyond the scope of social sciences
  • I do not believe that a stable financial system
    is feasible without broadly accepted social norms
    of personal responsibility business and
    accounting community
  • Our own (academic) community could not function
    without such norms
  • We should not expect that written codification of
    accounting standards to be sufficient to create
    order in the world of financial reporting, no
    matter how much resources and power is given to
    the Boards
  • They can only do so much. The rest is up to the
    community of accountants

51
The Challenge of Policy for Research
  • If research does not (at least ultimately) lead
    to better understanding, practice or policy, it
    will be ignored
  • The methods we have do not always take us in that
    direction
  • Stick to these methods (because that is what get
    published)
  • Or, move beyond the methods if they cannot
    enlighten us
  • Does research affect accounting policy for
    better?
  • Do standards improve financial reporting?
  • Do we have an honest assessment?
  • History suggests that improving financial
    reporting and governance is not an easy task

52
(No Transcript)
53
(No Transcript)
54
Whither Accounting Windows or Open Systems
  • Comfort vs. choice
  • Uniformity and stagnation vs. dynamic change
  • Predictability vs. some disorder
  • High prices or the advantages of technological
    progress
  • Financial reporting as an eco-system or a machine
    (garden or a building)
  • Huxley or Hayek
  • Nanny or personal responsibility
  • Where does the research community stand?
  • There are many life times of research agendas
    here if we are willing to consider them

55
References
  • Gode, D. K. and Shyam Sunder. 1993. Allocative
    Efficiency of Markets with Zero Intelligence
    Traders Market as a Partial Substitute for
    Individual Rationality. The Journal of Political
    Economy 101, no. 1 (February 1993) 119-137.
  • Sunder, Shyam. Determinants of Economic
    Interaction Behavior or Structure. Journal of
    Economic Interaction and Coordination 1, no. 1
    (May 2006) 21-32.
  • Sunder, Shyam. What Have We Learned from
    Experimental Finance? In Developments on
    Experimental Economics New Approaches to Solving
    Real-world Problems edited by Sobei H. Oda,
    91-100. Lecture Notes in Economics and
    Mathematical Systems 590. Berlin Springer, 2007.
  • William T. Baxter (Professor Emeritus, LSE), made
    many of these arguments over half-a-century ago
    (Recommendations on Accounting Theory in Baxter
    and Davidson, Studies in Accounting Theory, 1st
    edition).
  • Sunder, Shyam. 2010. Was Accounting a Root Cause
    of the Global Financial Crisis? Plenary Address
    to the Annual Meeting of the International
    Corporate Governance Network, Toronto, June 7.

56
Thank You!
  • Shyam.sunder_at_yale.edu
  • www.som.yale.edu/faculty/sunder
About PowerShow.com