28.2 Public Sector Data Sources and Reliability Verification

Public Sector Data Sources and Reliability Verification

Government analysis may draw from the general ledger and subledgers, budget and grants systems, procurement and payroll records, bank and payment files, asset systems, program case-management data, external registries, surveys, sensors, and open-data portals. Source authority does not itself prove reliability. Analysts document provenance and data ownership, reconcile record counts and control totals to an authoritative system, test completeness and valid time periods, inspect missing and duplicate values, validate field definitions and transformations, assess access and change controls, and corroborate high-risk fields with independent evidence. If limitations remain, the analyst narrows the conclusion or discloses the limitation rather than presenting an unsupported precise result.

Public financial managers and forensic auditors operate in an increasingly complex, data-rich environment. Modern governmental entities manage vast digital transaction ecosystems that capture financial, operational, demographic, and compliance information. Harnessing this information requires understanding core data sources and establishing strict data reliability protocols before conducting analytical or investigative procedures.

Core Sources of Governmental Financial and Operational Data

1. General Ledger & Enterprise Resource Planning (ERP) Systems: Provide the financial transaction backbone, recording all journal entries, chart-of-accounts hierarchies, budget allocations, commitments, encumbrances, and cash disbursements.
2. Human Resources & Payroll Databases: Contain employee master files, Social Security Numbers (SSNs), W-4 withholding data, hourly compensation rates, electronic timesheet approvals, overtime logs, and direct deposit routing numbers.
3. Procurement & Vendor Master Files: House vendor names, Employer Identification Numbers (EINs) or Taxpayer Identification Numbers (TINs), physical addresses, payment banking details, purchase orders, receiving slips, and historical contract awards.
4. Operational & Demographic Datasets: Encompass non-financial metrics critical for managerial performance evaluation, including census block demographic statistics, public school student average daily attendance (ADA) counts, fleet maintenance telematics, water utility smart-meter consumption records, and public safety dispatch logs.

Verifying Data Reliability: Integrity, Completeness, and Timeliness

Under Generally Accepted Government Auditing Standards (GAGAS / Yellow Book), auditors must evaluate the reliability of computer-processed data before using it to support audit findings. Flawed, incomplete, or manipulated data undermines both managerial decision-making and legal proceedings. Auditors verify data reliability through three primary dimensions:

• Completeness Verification: Ensuring that extracted datasets contain all valid transactions for the target audit period without unauthorized truncation. Technicians perform record count reconciliations between source ERP tables and working datasets, run gap tests on sequentially numbered documents (checks, purchase orders, warrants), and verify hash totals and cryptographic checksums (e.g., SHA-256) to confirm files were not altered during transmission.
• Data Integrity Checks: Validating internal data consistency. This involves executing field validity tests (e.g., confirming all SSNs contain exactly nine digits and no alpha characters), verifying format integrity (dates within valid fiscal ranges), testing mathematical cross-footing of ledger balances, and identifying orphan transaction records lacking valid foreign keys in vendor master tables.
• Timeliness and Timestamp Analysis: Examining transaction creation dates, batch posting dates, and approval timestamps. Discrepancies between system posting times and business hours (such as automated payments approved at 2:00 AM on Sunday) often indicate automated batch errors or deliberate control overrides.

---

The Four Tiers of Public Sector Data Analytics

Public sector analytics spans a capability continuum from historical reporting to automated cognitive decision modeling.

Descriptive (What happened?) 
   └──> Diagnostic (Why did it happen?) 
           └──> Predictive (What is likely to happen?) 
                   └──> Prescriptive (What action should be taken?)

1. Descriptive Analytics: Summarizing Historical Activity

Descriptive analytics synthesizes raw transaction data into understandable summaries, answering the fundamental question: What happened? Common examples include:

• Annual Comprehensive Financial Report (ACFR) summary tables
• Budget-to-actual variance reports by departmental cost center
• Expenditure breakdowns by object class (personnel, contractual services, supplies)
• Dashboard visualizations tracking total revenue collections by month

2. Diagnostic Analytics: Root-Cause Investigation

When descriptive reports reveal an anomaly—such as an unexpected budget overrun in overtime expenditures—diagnostic analytics drills into granular data to answer: Why did it happen? Techniques include:

• Drill-Down and Drill-Through Analysis: Navigating from total departmental expenditures down to individual employee timesheet entries.
• Correlation and Trend Decomposition: Comparing overtime spikes against emergency weather response logs, staffing vacancy rates, or seasonal recreation programs.
• Variance Analysis: Isolating price variances (wage rate increases) from volume variances (excessive labor hours expended).

3. Predictive Analytics: Forecasting Risk and Future Outcomes

Predictive analytics utilizes statistical modeling, regression analysis, and machine learning to project future occurrences, answering: What is likely to happen? Applications include:

• Revenue Forecasting: Multi-variable econometric regression estimating municipal sales tax receipts based on consumer confidence indices, local employment rates, and inflation indicators.
• Risk-Based Audit Selection: Predictive scoring algorithms evaluating public welfare claims (e.g., Medicaid, SNAP, unemployment insurance) to flag high-risk transactions for pre-payment audit review.
• Infrastructure Deterioration Modeling: Predicting water pipe burst probabilities or roadway pavement failure timelines using asset age, pipe material, and soil acidity data.

4. Prescriptive Analytics: Decision Optimization

Prescriptive analytics represents the most sophisticated analytical tier, combining optimization algorithms and simulation modeling to recommend optimal courses of action, answering: What should we do about it? Examples include:

• Municipal snow removal fleet route optimization models that dynamically reallocate plows to maximize roadway clearance speed while minimizing fuel expenditure.
• Debt portfolio optimization models that simulate interest rate sensitivity to determine the ideal blend of fixed-rate versus variable-rate debt under varying economic conditions.

Continuous Monitoring and Automated Exception Mining

Governmental internal audit shops and financial controllers increasingly deploy continuous monitoring scripts that run autonomously across ERP systems. Rather than performing retrospective, sample-based audits months after year-end, continuous auditing programs screen 100% of transactions against pre-programmed rules (e.g., flagging any vendor payment made to an employee address, or identifying invoices processed without a valid three-way match), alerting oversight personnel in near real-time.

---

Forensic Auditing Principles and the Fraud Triangle

Forensic auditing is the specialized application of accounting, auditing, data analysis, and investigative techniques to gather, analyze, and present financial evidence suitable for use in administrative hearings or criminal/civil courts of law. Forensic auditors do not merely verify conformity with GAAP; they actively search for intentional deception, asset misappropriation, and abuse of public office.

Distinguishing Fraud, Waste, and Abuse

In public program administration, oversight agencies distinguish among three related concepts:

1. Fraud: An intentional deception, false representation, or willful omission designed to deprive a government of assets or secure an unlawful personal advantage (e.g., submitting fraudulent invoices for undelivered medical supplies).
2. Waste: The thoughtless, careless, or extravagant expenditure of public resources without intentional deceit, resulting in unnecessary costs to taxpayers (e.g., purchasing expensive high-end computer servers that exceed operational requirements and sit unused in storage).
3. Abuse: Behavior that is deficient or improper when compared with behavior that a prudent person would consider reasonable and necessary, often involving the misuse of official authority or position for personal gain without necessarily violating a clear criminal statute (e.g., an agency director directing subordinates to perform personal errands during official government work hours).

The Donald Cressey Fraud Triangle in Government

Criminologist Donald Cressey established the Fraud Triangle, which posits that three conditions must be present for occupational fraud to occur:

                     PRESSURE / INCENTIVE
                       (Personal/Financial)
                              /   \
                             /     \
                            /       \
                           /  FRAUD  \
                          /  TRIANGLE \
                         /             \
                        /_______________\
       OPPORTUNITY                             RATIONALIZATION
(Weak Controls / Lack of SOD)           (Justification / Sense of Entitlement)

1. Pressure / Incentive: The motivation driving an individual to commit fraud. In government, pressures commonly stem from personal financial strain (gambling debts, medical bills, living beyond one's means) or organizational pressures (intense political demands to exhaust grant funding or meet artificial program targets to avoid budget reductions).
2. Opportunity: The structural opening that permits an employee to perpetrate and conceal fraud. In governmental entities, opportunity is created by weak internal controls, lax management oversight, single-user system access privileges, failure to perform independent reconciliations, and absence of adequate segregation of duties.

   Exam Focus: Opportunity is the Fraud Triangle element management can most directly influence and reduce through effective internal controls; no control system can guarantee complete elimination.

3. Rationalization: The psychological justification that allows an otherwise ethical individual to commit an unlawful act while maintaining an acceptable self-image. Common governmental rationalizations include: "The city underpays me compared to the private sector," "I am only borrowing the funds until my next paycheck," or "The government wastes millions anyway, so my small diversion doesn't hurt anyone."

---