Governmental Accounting

6 terms by jdonova4 

Ready to study?
Start with Flashcards

Create a new folder

Advertisement Upgrade to remove ads

Jerold Zimmerman
The Municipal Accounting Maze: An Analysis of Political Incentives
Journal of Accounting Research, 1977

(A) General questions of municipal accounting identified in this paper
(B) General Agency Theory
(1) Residual loss
(2) Agency Costs
(C) Three Hypotheses regarding Governmental Accounting
(D) Research Method
(1) City managers versus Mayors
(2) Empirical Hypotheses
(E) Results (2)

(A) General questions:
(1) What benefits does the mixing of cash and accrual accounting yield which offset the additional manipulation afforded the bureaucrat by this system?
(2) If legal statutes require cash accounting in certain instances, what prevents the city officials from also publishing supplementary financial statements based on full accrual accounting?
(3) Since municipal consolidations are probably less costly and no more confusing than corporate consolidations, why do most municipalities not produce them as supplementary statements?
(4) Why do municipal reporting and budgeting systems focus on the change in assets (inflows and outflows) on a fund by fund basis rather than the change in the municipality's wealth?
(5) Why is the extent and value of municipal owned assets rarely disclosed?

(B) Agent will not always act in the principal's best interests --- the principal can limit divergences by monitoring the agent and incurring monitoring costs
---- Agent will sometimes incur bonding costs to guarantee that he will not harm the principal
******* However it is unlikely that all issues will be solved by the monitoring and bonding costs
---- Residual loss - the value to the principal of the remaining divergence after monitoring and bonding costs are incurred
---- Agency costs - the sum of monitoring, bonding costs and the residual loss
---- Political agency costs: When such a political entrepreneur engages in activities (proposing, supporting, or opposing legislation) which further his career but which his constituency oppose, agency costs are generated

(C) Hypotheses:
(1) Supplementary financial statements based on full accrual accounting are not prepared because voters have little use for them
---- Under public property rights, ownership cannot be concentrated in individuals willing and able to monitor the costs and benefits of public property since there are restrictions on the sale of claims
---- Everyone owns an "equal" share of the public property and hence my incentive to monitor is greatly diminished
******* Public agents are less likely to be compensated for producing better accounting statements if the voters are unlikely to use them --- full accrual financial statements are there no prepared because voters have little incentive to monitor elected officials
(2) There is little demand for alternative sources of municipal accounting information
---- Constituencies being denied public services would like to know whether the refusal was because of a "true" lack of resources or because the politician was able to receive larger returns by allocating the resources to some other group
---- Making this differentiation is costly, and the constituency will incur the cost only if it can replace the elected official with a more sympathetic one
(3) Voters have little incentive to demand disclosure of the fair market value of the city's assets --- unlikely to be of use to creditors since general obligation debt is unsecured
---- If this information is useful it can also be provided outside the confines of the financial statements

(D) This section provides preliminary analysis of by examining disclosure differences between municipalities with a mayor as the chief executive officer and those with city managers
(1) Both forms involve an elected city council as a legislative body whose principle function are budget approval and rule making
---- City manager - non- elected administrator responsible for executing the policies of the council
******* An agency relationship exists between the city council and the appointed city manager ---- the council has incentives to monitor its appointed agent since collectively and individually council members' welfare is tied to the performance of their manager
---- Mayor as CEO - performs dignitary functions and/or is the presiding officer of the council ---- the mayor under the mayoral form is the chief executive officer of the city and in many cases also has legislative duties
******* Mayor is not the council's agent - council members and the mayor are elected directly by the people and, due to the separation of powers, they must cooperate, compromise, and make exchanges if services are to be provided

(2) Two Empirical Hypotheses
(i) Expect city councils that employ a managers are more likely than a city council-mayor organization to engage external auditors as a control device on the manager
(ii) Expect financial statements of city managers to be more detailed (more information regarding the various operations on the city)

(E) Results:
(1) Table 5 indicates that the usable mayoral cities are significantly less likely to be audited by and manager cities are more likely to engage large national auditors than mayoral cities
---- This is consistent with H1 --- to the extent that a large CPA firm provides more protection against fraud than a small CPA firm, large CPA firms will be engaged more frequently by manager councils than by mayoral councils
(2) Table 6 - usable manager reports are longer and contain more exhibits than usable mayoral reports
---- Difference in the mean length of exhibits is significant at the 10% level using a one-tailed test

Nolan Kido, Reining Petacchi, and Joseph Weber
The Influence of Elections on the Accounting Choices of Governmental Entities
Journal of Accounting Research, 2012

(A) Research question
(B) Motivation and Hypotheses (3)
(C) Research Method
(1) Governmental accruals examined
(2) Model of discretionary accruals
(3) Proxies for cross sectional tests (3)
(4) Main test
(D) Result

(A) What is the relation between an indicator variable for election years and the residual of a regression of changes in the compensated absences liability on employee payroll, changes in the state's employee benefit policies, state GDP, the residual from a regression of employee compensation on the number of employees, and firm size?

(B) When incumbents run for reelection the fiscal performance of the governmental entity during their term in an important part of the election platform
---- Provides incentives for politicians to make accounting choices that imply an overall healthy financial condition of the state
---- States with rigorous budget restrictions have stronger incentives to employee accounting practices to ensure expenditures do not exceed revenues as an altnerative to cutting expenditures or increasing taxes
---- States with weak restrictions can use other fiscal techniques to carry deficits forward and thus have fewer incentives to use accounting manipulations
---- States that are expected to have relatively poorer fiscal performance will have stronger incentives to use accounting direction to appear healthier - expect that states in poor health have stronger incentives to engage in accounting manipulations to attract votes
---- Federal law requires states to conduct annual audits of their financial statements and internal control systems - position may be independently elected, appointed by the legislature, appointed by the governor, or jointly appointed
******* Auditors who are directly elected by citizens are more independent and are more likely to offer higher audit quality - more likely to hinder a politicians' ability to adopt favorable accounting treatment during an election year

Hypotheses:
(1) Discretionary components of the liabilities associated with compensated absences and unfunded pension will be smaller (more negative) in the fiscal year prior to an election
(2) States in poor financial health or states with strict budget restrictions have stronger incentives to use discretion to reduce their liabilities
(3) Incumbent parties' ability to manipulation accounting information in an election year will be smaller for sates with more independent state audit agencies and more independent state controllers

(C) Research Method:
(1) State's work force liability accruals
(i) Compensated absences liability
(ii) Unfunded pension liabilities

(2) Discretionary accruals:
(i) Model of discretionary change in the compensated absences liability: model the change in this account balance as a function of the change in the payroll for the state's full-time employees, changes in the state's employee benefit policies, and a series of control variables
---- The residuals from this regression reflect the unexplained change, and serve as our measure of the "discretionary" portion of this liability

∆CA = α₀ + α₁∆Payroll + α₂Policy + α₃Unexpected Compensation + α₄GDP Growth + α₅Size + ε

(3) Proxies for cross sectional tests:
(i) Proxy for independence of state auditors and controllers - whether state auditors (controllers) are elected by citizens or appointed by state governments as a proxy for independence
---- Auditors who are elected are more likely to conduct high quality audits and less likely to afford the state government discretion
---- Expect elected controllers to be more independent and less likely to prepared biased financial reports

(ii) Measure of the restrictiveness of balance budget requirements following Hou and Smith (2006) --- five most restrictive budgetary rules
---- Define state as weak governance (WeakBBR = 1) if it does not have any of the ant deficit rules listed below:

(iii) Measure of financial health: ratio of forecasted year-end total balances to forecasted generated expenditures - scaled measures of general account surpluses are common measures used to evaluate a state's fiscal health


(4) Main regression:

Discretionary ∆CA = β₀ + β₁Election + β₂ElectionWeak BBR + β₃ElectionFinancial Health + β₄ElectionAuditor Independence + β₅ElectionPrepare Independence + β₆Weak BBR + β₇Financial Health + β₈Auditor Independence + β₉Prepare Independence + ε

---- Main result: expect a negative coefficient on the Election indicator (β₁) consistent with the idea that states record smaller compensated absences liability balances in election years
---- Also expect the coefficients on each of the interaction variables will be positive because accounting manipulation
on compensated absences will be mitigated by weak balance budget requirements, strong financial health, high audit quality, and high preparer independence

(D) Overall results indicate that in an election year recorded liability associated with compensated absences and unfunded pension liabilities are both systematically lower --- suggests that state governments manipulation accounting numbers to present a healthier financial picture in an election year
---- Find that the discretionary component of the change in compensated absences is smaller in the fiscal year right before a gubernatorial election - compared to non-election years, state governments on average abnormally accumulate $23.9 million less in compensated absences in an election year

(1) Figure 1: Plot the discretionary component of changes in the compensated absences liability in time where t=0 is the fiscal year right before the election
---- We see a dip in the discretionary components of changes in the compensated absence liability during the election year ---- provides support for H1 regarding the opportunistic accrual balance of the compensated absences liability prior to a gubernatorial election

(2) Table 3: Regression results testing the impact of election years on the discretionary changes in the compensated absences liability
---- Negative and significant coefficient on the Election indicator variable suggests that the balance in the liability is significantly lower in election years
******* Comparing the significant coefficient on Election with the intercept indicates t that changes in the discretionary compensated balances liability decreases almost four time an election year relative to a non-election year
---- The positive and significant coefficients on the interaction variables (one-tailed tests) are consistent with the hypotheses that the incentives to manipulate the discretionary changes in the compensated absences liability are mitigated if the state has weak blanked budget restrictions, is in greater financial health, has an independent elected auditor, or has an independent elected controller
******* Sum of the coefficients on Election and WeakBBR is not significantly different from zero --- suggests that since during election years states are less likely to cut pending or increase taxes, states with strict anti-deficit provisions are under more pressure to meet the budget and thus have an even stronger incentive to manage accounting numbers
******* Sums of the coefficients for independent auditors and independent prepares are both statistically smaller than zero, which suggests that independent state auditors and controllers cannot fully curtail the accounting management of the accrual

(3) Table 4: Robustness tests ensuring that results are not driven by concerns about employees losing their leave so they use it in the election year --- should observe mitigating effect from strong union membership and public sector collective bargaining power if this is the case
******* Column 1 and 2 - results do not support the altnerative explanation as the interaction of Election and the proxies for union bargaining power are insignificant
******* Column 3 - interaction of the bargaining power proxies and the Election indicator variable is insignificant --- not consistent with altnerative explanation
******* Column 4 - insignificant coefficient on the interaction of Election with the proxy for high state level employment rules out the altnerative explanation that states without high state employment do not want to see their tax dollars being used to compensate state employees

(4) Table 5: The effect of election years on the discretionary unfunded pension liabilities
---- Negative and significant coefficient on the Election indicator variable indicates that the balance is the unfunded pension liability is abnormal small in an election year
******* Positive and significant coefficient on the interaction of Election with Financial Health, and Independence variables are consistent with previous results that the abnormal use of the discretion is mitigated if the state is in greater financial health, has an independent elected auditor, or has an independent elected controller
---- Overall this table provides similar results to previous analysis and mitigates concerns that the main findings are attributable to omitted variables

Robert Ingram
Economic Incentives and the Choice of State Government Accounting Practices
Journal of Accounting Research, 1994

(A) Research question
(B) Motivation and Hypothesis
(C) Research Method
(1) Sample selection
(2) Three Dependent Variables
(3) Key economic factors considered
(D) Result

(A) What is the relation between the 12 (and 8) practice index summing the number of accounting practices out of the 12 recommended by GAAP (and an indicator variable = 1 if the state had 5 or more of the 12 practices (high information state), and 0 otherwise) and the following state economic factors: the presence of special interest groups with incentives to monitor, the extent of debt outstanding, own revenue Per capital, governmental salaries, the requirement that an auditor be a CPA, and state population?

(B) Governmental accounting relies on the use of fund accounting to limit the set of options available to politicians
---- Coalitions of voters might have greater incentives to monitor political behavior since the potential result of their collective vote may be important for the election
---- Administrators of the accounting systems and the auditor are well positioned to influence accounting practices- these individuals can monitor the activities of other officials and alter the disclosure system itself
---- If the bond market recognizes conformity as a signal of management quality, there may be incentives to adopt GAAP to signal management quality and reduce interest costs

Hypothesis: Expect all economic variables will be positively associated with the level of accounting disclosure provided by states

(C) This study examines the association between economics factors and cross-sectional variations in accounting practice of state governments

(1) Recent survey by the Council on State Governments (CSG) in 1980 provides summary of major accounting and reporting practices for individual state governments --- characterizes both the general status of state government accounting and the diversity of accounting practices observed across states
---- These survey indicate that accounting practices vary considerably across governments and departure from GAAP was common - no attempt there to evaluate whether these practices are good or bad but instead the perspective is taken that states and politicians that employ practices that are in their own corporate or individual best interests
---- In general more complete reporting is directly associated with the extent o which reporting practices conform to GAAP --- assume that conformance to GAAP can be interpreted in this setting as representing more information disclosure

(2) Three dependent variables:
(i) 12-practice index summing the number of accounting practices out of the 12 recommended by GAAP in Table 3 for each state
(ii) 8-practice index summing the number of accounting practices out of the 12 recommended by GAAP in Table 3 for each state
(iii) Indicator variable = 1 if the state had 5 or more of the 12 practices (high information state), and 0 otherwise (low information state for firms with 4 or less practices)

(3) Major independent variables:
(i) Special interest groups that might have incentives to monitor
(ii) Extent of debt outstanding = indicator for internal incentives to minimize interest costs (which can be facilitated by increased disclosure)
(iii) Own Revenue Per capita = general revenues other than intergovernmental transfers --- proxy for fiscal resources available to cover the costs of accounting changes
---- May also represent the need for additional information because of the large resource base available
(iv) Salaries and the requirement that an auditor be a CPA - proxy for managerial professionalism
(v) State population = proxy for the complexity of government
******* Performs principal components analysis on the independent variables in the model to lower the dimensionality and mitigate multi-collinearity problems

(D) Overall results are consistent with the idea that the extent of state government accounting disclosure is affected more by the demands of and needs for information exhibited by constituents and administrators than recommendations and accounting standards

(1) Table 7: Correlation matrix between each independent variable and an accounting practice index based on the 12 practices used in Table 4 and an alternative index based on the first 8 practices in Table 4
---- Fifteen of the 16 correlations in each column are positive --- newspaper circulation is negative
---- Each variable is expected to be positively associated with the quantity of accounting information --- these results are consistent with expectations
******* Six product moment correlations were significant at the < .10 level ---- and eight rank-order correlations were significant for the 12 practice index and 7 for the 8-pracice index

(2) Table 9: Testing the effect of economic factors on the amount of accounting information provided by state governments
---- Results indicate that the Newspaper Circulation coefficients are negative --- indicates that greater press strength results in less accounting information
******* This may be consistent with the idea that the press may be a cost-effective substitute for accounting disclosure
---- Positive coefficient on all remaining variables is consistent with expectations that higher economic values in these variables increases the amount of disclosure in accounting information
******* Most significant factors: urbanization, political competition, governor (which indicates that more accounting information is provided by governor appointed accounting administrators systems than by other forms of selection)
---- Larger incentives on the part of voters to monitor government finances, larger internal incentives to monitor by appointed administrators and larger incentives to reduce debt costs combined with higher paid administrative personnel are all associated with more accounting information
******* Governor appointed administrators and legislature appointed auditors are associated with higher information production than elected officials ---- possible that these findings are that elective and appointive processes result in different types of administrative personnel with regard to their training and understanding of accounting systems and their incentives to monitor government finances

William Baber and Angela Gore
Consequences of GAAP Disclosure Regulation: Evidence from Municipal Debt Issues
The Accounting Review, 2008

(A) Research question
(B) Motivation and Hypotheses (3)
(C) Research Method
(1) Sample Selection
(2) Three Main Tests of Hypotheses
(D) Result

(A) What is the relation between an indicator variable for GAAP-mandated disclosure states and (i) the percentage of private debt in the municipality's total debt and (ii) the rate of return that equates the proceeds of the issue with the present value of interest and principal payments (debt costs)?

(B) If tangible benefits of GAAP regulation do not accrue then decisions to impose GAAP reporting apparently contradict the public interest
---- If state GAAP requirements are associated with lower debt financing costs then the characteristics of debt issues should differ between GAAP states and non-GAAP states
******* If GAAP disclosure regulation reduces the costs of contracting between municipal borrowers and lenders expect greater use of debt by municipalities in GAAP than in non-GAAP states
---- Disclosure regulation can also reduce transaction costs and influence liquidity in lending markets - if disclosure improves market liquidity then decisions to disclose voluntarily fail to incorporate public benefits that accrue to other lenders from liquid capital markets
---- Disclosure requirements can also increase agency costs - disclosure of proprietary information in the private sector can undermine an organizations' position with respect to competitors or contracting with other external parties
---- Managers can respond to disclosure regulation by reducing reliance on other less costly mechanism that could be implemented voluntarily in an unregulated environment
******* In general there are higher transaction costs and greater information asymmetry in municipal debt markets versus the corporate debt markets

Hypotheses:
(1) The issuance of municipal debt is independent of whether GAAP is required by the state government
(2) The relative use of private (versus public) municipal debt financing is independent of whether GAAP is required by the state government
(3) The cost of municipal debt financing is independent of whether GAAP is required by the state government

(C) Research Method
(1) Sample selection: identify 15 states that require municipalities to prepare annual GAAP based financial statements (regardless of whether they issue debt) and 10 states that do not require annual financial reporting by municipalities: to be classified in a GAAP state, the requirement must be imposed by state law

(2) Test the relation between GAAP disclosure regulation and the use of municipal debt financing:

DEBTUSE = α₀ + α₁GAAP State + αControls

---- Two dependent variables:
(i) Dependent variable A = indicator variable = 1 if the firm issued private or public debt, 0 otherwise period from 1995-2002
(ii) Dependent variable B = Continuous variable = total new debt per capita issued during the period from 1995-2002

---- Coefficient on GAAP State (α₁) > 0 indicates municipal debt use is greater for GAAP than non-GAAP states

(3) Test the impact of state GAAP disclosure regulation on the choice of public versus private debt:

Private Debt / Total Debt = β₀ + β₁GAAP State + βControls

---- Coefficient on GAAP State (β₁) < 0 indicates that municipalities in GAAP states are more likely to public debt versus non-GAAP states

(4) Test the impact of GAAP disclosure regulation on the firm's interest costs

Interest Costs = γ₀ + γ₁GAAP State + γ₂INDEX + γControls

---- Dependent variable: True mean interest costs (TIC) = rate of return that equate the proceeds of the issue with the present value of interest and principal payments
---- Coefficient on GAAP State (γ₁) < 0 indicates that interest costs are lower for municipalities in GAAP states versus unregulated states

(D) Overall results indicate great use of public debt (versus private debt) by municipalities in GAAP states versus non-GAAP states, which suggests that reductions in municipal debt costs associated with state mandated GAAP are greater for public than for private debt issues

(1) Table 3: Testing the impact of GAAP mandated disclosure on the use of debt, both private and public
---- The coefificient on the GAAP state indicator variable is insignificant, which is inconsistent with rejecting the null hypothesis of H1 that GAAP disclosure has no impact on the access of debt (public and private)
******* No evidence that GAAP disclosure regulation significantly influences debt issuance beyond what is achieved through voluntary disclosure and other state-imposed mechanisms

(2) Table 4: The impact of mandatory GAAP disclosure on the ratio of private debt to total debt issued during the period 1995-2002
---- Negative and significant coefficient on the GAAP state indicator variable indicates that GAAP states with required disclosure are significantly more likely to access the public debt markets for financing versus non-GAAP states
******* This result is consistent with rejecting H2 and suggests that municipalities in GAAP states are less likely to place debt issues privately (more likely to issue public debt)

(3) Table 5: Testing the effect of state mandated GAAP disclosure on the municipalities' interest costs
---- Negative and significant coefficient on the GAAP state indicator variable suggests that firms in states with GAAP mandated disclosure obtain significantly lower debt costs versus non-GAAP disclosure states
******* Column 1 - mean true interest costs in GAAP states are 16 basis points lower than in unregulated states

(4) Table 6: The effect of GAAP mandated disclosure on debt costs in Nevada and Montana after new adoption of the regulation
---- Panel A: Negative and significant coefficient on the indicator variable for issues post regulation indicates that debt costs are lower in these two states following the adoption of state regulation requirement mandatory disclosure
******* Additionally, the results for comparisons of the regulation year with the year prior to the regulation year are significantly different

Elizabeth Plummer, Paul Hutchinson, and Terry Patton
GASB No. 34's governmental financial reporting model: Evidence of its information relevance
The Accounting Review, 2007

(A) Research question
(B) Motivation and Hypotheses (3)
(C) Research Method
(1) Measure of default risk
(2) Government-Wide Statements variables
(3) Governmental Funds Statements variables
(4) Model to test H1-H3
(D) Result

(A) What is the relation between school district debt ratings and (i) government wide financial position and financial performance (ii) individual components of the government wide financial position and financial performance, and (iii) the funds statements components of financial position and financial performance?
---- Additionally which of these independent variables better explains the dependent variable of debt ratings?


(B) GASB 34 requires governments to publish governmental funds statements prepared on a modified accrual basis and to publish government wide statements prepared on an accrual basis --- makes government financial reporting more closely resemble corporate financial reporting
---- Four reasons to expect the accrual-basis statements are likely to be more relevant than the funds statements:
(i) Relative to expenditures on a modified accrual basis, accrued expenses reported on the Statement of Activities should provide a better measure of the economic costs of providing government services and whether a government's revenues are sufficient to cover its operating costs
(ii) Statement of Net Assets should provide better information than the funds' Balance Sheet about the long-term viability of a government's revenue versus cost structure
(iii) GASB No. 34 also requires that total net assets be displayed in three separate components: invested in capital assets, net of related debt (INVCAP) restricted net assets (RNA); and unrestricted net assets (UNA)
(iv) GASB No. 34 requires that capital assets be reported on the Statement of Net Assets and subsequently depreciated

---- Two reasons why accrual basis statements will not provide incremental information over the funds statements:
(i) Prior literature suggests that the relative superiority of earnings is largely attributable to working capital accruals and not long-term accruals. The governmental funds statements are based on a modified working capital model
******* If the improvement in accrual-basis earnings is primarily attributable to working capital accruals, then the government-wide Statement of Activities is not likely to provide incremental information over the funds' Statement of Changes
(ii) Market demand governmental financial statements is likely to be lower than the market demand for public companies' financial statements (Pinnuck and Potter 2006).9 If lower demand results in lower quality, then the accrual-basis statements may not provide incremental information over the governmental funds statements

Hypotheses:
(1) There is a significant association between the accounting information provided by government-wide financial statements and school districts' default risk
(2) There is a significant association between the separate components of total net assets (invested in capital assets, net of related debt (INVCAP) restricted net assets (RNA); and unrestricted net assets (UNA) and school districts' default risk
(3) Relative to the fund statements, government-wide financial statements provide incremental information about school districts' default risk

(C) Research Method
(1) Measure of default risk that is based on a school district's financial condition and that is observable is the district's underlying debt rating. This rating is based on an agency's assessment of a particular debt issue's credit quality absent credit enhancement
---- Values between 1 and 14, where larger values correspond to a less favorable debt rating

(2) Government-Wide Statements Variables:
(i) POS(GW) =  financial position, computed as total net assets, divided by total revenues;
(ii) PERF(GW) = financial performance, computed as total revenues minus total expenses, divided by total revenues;
(iii) INVCAP = invested in capital assets, net of related debt, divided by total revenues;
(iv) UNA = unrestricted net assets / total revenues;
(v) RNA_DEBT = net assets restricted for debt purposes, divided by total revenues;
(vi) RNA_OTHER  = net assets restricted for non-debt purposes, divided by total revenues
(vii) CL(GW) = current liabilities, not including the current portion of long-term bonds, divided by total revenues

(3) Governmental Funds Statements Variables
(i) POS(FUND) = financial position, computed as total fund
balance, divided by total fund revenues;
(ii) PERF(FUND) = (total fund revenues - expenditures) / total fund revenues
(iii) CL(FUND) = fund liabilities / total fund revenues

(4) Test H1: Ordered logit model to test the ability of government wide financial statements to explain default risk (relevance):

RATING = β₀ + β₁POS(GW)Small + β₂POS(GW)Large + β₃PERF(GW)Neg + β₄PERF(GW)Pos + βControls + e

---- Small = an indicator variable equal to 1 if POS(GW) is smaller than the median, and 0 otherwise
---- Large = an indicator variable equal to 1 if POS(GW) is larger than the median, and 0 otherwise
---- Neg = an indicator variable equal to 1 if PERF(GW) is negative, and 0 otherwise
---- Pos = an indicator variable equal to 1 if PERF(GW) is positive, and 0 otherwise

****** Expect negative coefficient on the POS(GW)Small, POS(GW)Large, and β₃PERF(GW)Neg variables
****** Expect positive coefficient on PERF(GW)Pos

(5) Test H2: Ordered logit model to test the ability of individual components of government wide financial statements to explain default risk (relevance):

RATING = α₀ + α₁INVCAPSmall + α₂INVCAPLarge + α₃UNASmall + α₄UNALarge + α₅RNA_DEBT + α₆RNA_OTHER + α₇PERF(GW)Neg + α₈PERF(GW)Pos + α₉CL(GW) + αControls + e

****** Expect negative coefficients on the INVCAPSmall, INVCAPLarge, UNASmall, and UNA*Large variables

(6) Test H3: Ordered logit model to test the ability of governmental funds statements to explain default risk (relevance):

RATING = γ₀ + γ₁POS(FUND)Small + γ₂POS(FUND)Large + γ₃PERF(FUND)Neg + γ₄PERF(FUND)Pos + γ₅CL(FUND) + γControls + e

---- Small = an indicator variable equal to 1 if POS(FUND) is smaller than the median, and 0 otherwise
---- Large = an indicator variable equal to 1 if POS(FUND) is larger than the median, and 0 otherwise
---- Neg = an indicator variable equal to 1 if PERF(FUND) is negative, and 0 otherwise
---- Pos = an indicator variable equal to 1 if PERF(FUND) is positive, and 0 otherwise

****** Expect negative coefficient on the POS(FUND)Small, POS(FUND)Large, and PERF(FUND)Neg variables
****** Expect positive coefficient on the PERF(FUND)Pos and CL(FUND) variables

(D) Overall results indicate that GASB 34's Statement of Net Assets (i.e. Balance Sheet) provides information relevant for assessing default risk and this information is incremental to that provided by the governmental funds statements
---- However GASB 34's Statement of Activities (i.e. income statement) does not provide information relevant for assessing default risk --- accrual earnings are not more relevant than the modified accrual earnings

(1) Table 5: The ability of the Government Wide financial statements and the components of the government wide financial statements to model default risk using an ordinal logit model
---- Equation (1) : the relevance of government wide statements for explaining default risk - results provide partial support for H1 that government wide statements provide relevant information for default risk:
****** Negative and significant coefficients on the interactions of POS(GW)Small and POS(GW)*Large are consistent with H1 --- suggests that default risk increases as the government-wide net assets increases
******* Coefficients on PERF(GW) and CL(GW) are not significantly different from zero, which is inconsistent with H1 --- this suggests that these government wide variables do not provide incremental information about debt ratings
******* Model pseudo-R square is relatively high, suggesting that the model does a relatively good job of explaining debt ratings

---- Equation (2): the relevance of the components of government wide total assets in explaining default risk ---- overall results support H2 and suggest that the components of total assets are informative about the underlying debt ratings
****** Significantly negative coefficients on INVCAPSmall and INVCAPLarge and UNASmall and UNA*Large suggests that default risk decreases as INVCAP and UNA increase

(2) Table 6: The ability of fund statements to reflect default risk of the school district --- general evidence on the informativeness of the funds statements accounts
---- Results are generally consistent with expectations that funds statements accounts provide information relevant to the explanation of default risk
****** Negative and significant coefficient on deficit spending PERF(FUND)Neg and positive and significant coefficient on surplus spending PERF(FUND)*Pos indicates that default risk decreases as current year deficits become less negative and that default risk increase as current year surpluses increase
******* Significantly positive coefficient on CL(FUND) indicates that default risk increases as liabilities on the funds statements increase
---- The pseudo r-square of the model is 71.8% which suggests that the model generally does a good job at estimating default rates

(3) Table 7: Test of H3 -- the relative ability of the government wide and funds financial statements to explain default risk --- compare the information content of the government wide statements with the funds statements
---- Estimates (1) the funds model, (2) the government wide model, and (3) the expanded model which includes control variables, funds variables, and government wide variables
******* Examine the incremental information content of government wide variable relative to the funds variables using the Likelihood Ratio chi-square test

---- Results are consistent with the results in Tables 5 and 6 - POS(GW), PERF(FUND), and CL(FUND) provide relevant information for explaining deb ratings but POS(FUND), PERF(GW), and CL(GW) do not
******* Likelihood Ratio Test - if ΔΧ2 exceeds the critical value - conclude that the government wide variable significantly improve the model and provide incremental information for explaining debt ratings

---- Panel A: POS - ΔΧ2 = 6.84 is significant at the 5% level consistent with H3 --- suggests that the government wide measures of financial position provides incremental information over the funds components of position in explaining the districts' debt ratings
---- Panel B: PERF- ΔΧ2 = 0.42 is insignificant which is inconsistent with H3 --- suggests that the government wide measures of financial performance do not provide incremental information over the funds components
---- Panel C: CL - ΔΧ2 = 2.41 is insignificant which is inconsistent with H3 --- suggests that the government wide measures of current liabilities do not provide incremental information over the funds components

Nicholas Apostolou, Gary Giroux and Robert Welker
The Information Content of Municipal Spending Rate Data
Journal of Accounting Research, 1985

(A) Research question
(B) Motivation
(C) Research Method
(D) Results

(A) What is the relation between municipal net interest cost and the spending rate (revenue / expenditures), and how does this relation vary according to deficit and surplus spending?

(B) The objective of this study was to further investigate whether the spending rates of municipalities, computed using publicly available accounting data, contain information useful to bond investors in assessing the relative riskiness of new municipal bond issues

(C) Non-linear regression model used to test the association between net interest cost (dependent variable) and spending rate and an indicator variable for surplus or deficit spending

(D) Overall results indicate that that reported spending rates are associated with net interest cost but the association is not monotonic, with a change in direction of the slope occurring at the balanced-budget point (revenues equal expenditures)
---- Findings suggest that the rate of surplus spending and the rate of deficit spending are both positively related to net interest cost of long-term municipal bonds at approximately the same rate of change

(1) Table 1: Estimates of the three regression model indicated above based on the natural log transformation of the three models
---- Liner SR Model - makes no distinction between surplus spending and deficit spending - no significant relation between spending rate and net interest cost
---- Nonlinear SR Model - assumes the relationship between surplus spending and net interest cost differs from the relationship between deficit spending and net interest cost (indicator variable used to assess this difference)
******* F-test rejects the null hypothesis that the model is not well specified at the < .05 level --- indicates that the spending rate is related to net interest cost but that the function relation is dependent on whether the municipality has a surplus or deficit spending rate
******* No significant difference (F=.25) between the slopes --- indicates that an increase in surplus of deficit spending increase net interest cost at approximately the same rate of change

Please allow access to your computer’s microphone to use Voice Recording.

Having trouble? Click here for help.

We can’t access your microphone!

Click the icon above to update your browser permissions above and try again

Example:

Reload the page to try again!

Reload

Press Cmd-0 to reset your zoom

Press Ctrl-0 to reset your zoom

It looks like your browser might be zoomed in or out. Your browser needs to be zoomed to a normal size to record audio.

Please upgrade Flash or install Chrome
to use Voice Recording.

For more help, see our troubleshooting page.

Your microphone is muted

For help fixing this issue, see this FAQ.

Star this term

You can study starred terms together

NEW! Voice Recording

Create Set