Estimate Accruals and Earnings Quality
High accruals may associate with lower subsequent earnings (i.e., low earnings persistence of accruals; please see Portland General Electric Co ). This page gives you a quick way to compute the accruals of Portland General Electric Co and identify the quality of the firm’s reported earnings, which will help you to evaluate the firm’s future performance.
A spreadsheet module
A spreadsheet module is provided for Portland General Electric Co to calculate its accruals, error terms, and the standard deviation of the errors . Please click AccrualCalculator to download the module. The definition of accruals is provided in Appendix A of this web page. A larger error term and higher standard deviation of the error terms means poorer quality of accruals and reported earnings (see Appendix B for detailed introduction).
To view the results, please input the 2019 financial information based on the firm’s just announced financial report. That is, update the information in the green cells as shown below. Information for other years is already included.
|
| [1] | [2] | [3] | [4] | [5] | [6] | [7] |
|
| million $ | ||||||
| Year (t) | Current assets (CA) | Current liabilities (CL) | Cash and short-term investments (Cash) | Debts in current liabilities (Debt) | Total assets (TA) | Cash flow from operating activities (CF) | Earnings (operating income after depreciation) (E) |
|
| Source: balance sheet | Source: balance sheet | Source: balance sheet | Source: balance sheet | Source: balance sheet | Source: Statement of Cash Flow | Source: Income Statement |
|
|
|
|
|
|
|
|
|
| 2019 |
|
|
|
|
|
|
|
After inputting the information, you will see the results on the bottom right-hand side of the sheet (see example below), including the firm’s 2019 accruals, average accruals over the last ten years, 2019 accruals error, and the standard deviation of accruals errors over the last ten years. Moreover, the position of these statistics in the firm’s industry and market are also presented. Q1 indicates a very low level (1st quartile, or the bottom 25%), Q2 indicates a low level (2nd quartile), Q3 indicates a high level (3rd quartile), and Q4 indicates a very high level (4th quartile or top 25%).
| The firm's accrual level and quality | The firm's position in the industry or market. Q1-very low (1st quarter, bottom 25%), Q2-low (2nd quarter), Q3-high (3rd quarter), and Q4-very high (4th quarter) | |||||
| industry | market | |||||
| Accrual level | Accrual in 2019 | 0.085 | Q4 | Q4 | ||
| Average accruals over the past 10 years | -0.007 | Q1 | Q2 | |||
|
|
|
|
|
|
|
|
|
| ||||||
| Accrual quality | Accrual errors in 2019 | 0.020 | Q4 | Q4 | ||
| Standard deviation of errors over the past 10 years | 0.014 | Q1 | Q2 | |||
Appendix A: Measuring accruals and cash flows
The difference between accounting earnings and underlying cash flows is called accruals. The key element of accruals is the change in firm net current operating assets (current assets – current liabilities), often called working-capital accruals. Cash is excluded from current assets because accruals arise due to changes in non-cash assets. Debts in current liabilities are also excluded from current liabilities because they belong to financial activities rather than operating activities. Specifically, working-capital accruals can be written as follows:
Working-capital accruals = change in current assets (excluding cash) – change in current liabilities (excluding debts in current liabilities) (1)
For example, in 2019, company X had a $60 million increase in current assets (excluding cash) and a $20 million increase in current liabilities (excluding debts in current liabilities). There were no changes in other operating assets. Thus, the firm’s working-capital accruals were 60 – 20 = $40 (million).
Accruals also include the depreciation of fixed assets (i.e., non-current operating assets, such as property, plant, and equipment). Depreciation enters into accruals with a negative sign to recognize a decline in future benefits from fixed assets. Accruals in this context are called total accruals, which can be described as follows:
Total accruals = change in current assets (excluding cash) – change in current liabilities (excluding debts in current liabilities) – depreciation (2)
Suppose that company X had $10 million of depreciation expenses in 2019, the firm’s total accruals were 60 – 20 – 10 = $30 (million).
The difference between earnings (i.e., operating income after depreciation) and total accruals is called cash flow from operations, which represents the cash generated by a company’s normal business operations. This relationship can be described as follows:
Cash flows from operations = earnings – total accruals (3)
If company X had $100 million in net income in 2019, then the company’s cash flow from operations was 100 – 30 = $70 (million). This suggests that the company recorded $100 million of accounting earnings but only realized $70 million in cash.
Appendix B: Measuring the quality of accruals and earnings
High accruals are associated with lower subsequent earnings (i.e., low earnings persistence of accruals; please see Portland General Electric Co ) because the anticipated future benefits have not been realized. Therefore, the low persistence effect of accruals is likely to be stronger when the mapping between accruals and cash flow realization is weak; that is, when accrual estimation errors are large. To evaluate the earnings persistence of accruals, we can regress accruals on cash flows in historical, contemporary, and future periods. The accruals that are unrelated to these cash flows measure the mismapping between accruals and actual cash flow.
Specifically, working-capital accruals for a specific firm in year t can be matched with the firm’s cash flow from operating activities in year t–1, t, and t+1 and then regression analysis can be performed. In each year, there is an estimated error term, which is the difference between the estimated and actual accruals. The error term thus represents the mismapping between accruals that year and cash flow realization. A larger error means greater mismapping. The standard deviation of the time-series errors measures a firm’s earnings or accruals quality. A higher standard deviation means poorer quality. Details of the regression model are provided below.
Regression models
To measure earnings and accrual quality , we can regress accruals on cash flow realization using the following model . 1
WCA t = a 0 + a 1 * CFO t-1 + a 2 * CFO t + a 3 * CFO t+1 + e t
This model can be estimated at the firm level. That is, an OLS regression can be run for a firm using its current and past annual data (requiring a minimum of five observations).1
The error term e t captures the extent to which accruals map on to realized cash flow. A high value indicates poor cash flow realization. The standard deviation of the time-series errors can be used as a measure of accruals and earnings quality. A high standard deviation indicates low quality.
If accruals are defined to incorporate depreciation, then we can estimate accruals quality using the following model.2
TA t = b 0 + b 1 * ( 1⁄Assets t-1 )+ b 2 * ∆Sales t + b 3 * PPE t + b 4 * ROA t + e t
TA t : Accruals (defined by Eq. 2) in year t. Specifically, TA t = ( Change in current assets t - Change in cash t - Change in current liabilities t + Change in short_term debt t - Depreciation t )/ Total assets t-1 .
Assets t-1 : Total assets in year t-1.
∆Sales t : Change in sales in year t, scaled by total assets in year t-1.
PPE t : Net property, plant, and equipment in year t, scaled by total assets in year t-1.
ROA t : Net income scaled by total assets in year t.
The error term e t captures the quality of accruals. A higher level and greater variation of e t indicate greater discretion in accruals estimation and thus lower earnings quality.
References
1Dechow, P. M. & Dichev, I. D. The quality of accruals and earnings: The role of accrual estimation errors. The Accounting Review 77, 35-59 (2002).
2Kothari, S. P., Leone, A. J. & Wasley, C. E. Performance matched discretionary accrual measures. Journal of Accounting and Economics 39, 163-197 (2005).