Chapter 4 - Accounting in Government

Return to toc

Introduction

Governments are accountable to their electorates. Formal rules of accountability are enshrined in legislation. Government business enterprises are generally subject to at least part of the provisions of relevant companies legislation and to national competition policy.

A major part of public sector reform has been to change the focus of accountability. Historically, governments were accountable mainly for their adherence with budgetary appropriation; did they spend what was appropriated, and did they spend the funds on the correct items? The emphasis was on precision and compliance with line item appropriation, correctly called an input based obsession. If a department was appropriated $294 603 for postage, then woe betide the hapless bureaucrat who misappropriated these funds on telephone, or who underspent them by ten percent.

The general directions of reforms has been to make governments more accountable for outputs (what did they produce?) and with outcomes (what did these outputs achieve?). These are generally known as effectiveness indicators. Also there is concern with efficiency (was this output achieved at lowest possible cost?).

It is natural, therefore, that governments should turn to the private sector for accounting models. Private sector accounting standards are designed to ensure managers can show their stakeholders how well they are managing the resources under their trust. Corporate efficiency can usually be encapsulated in a few simple financial ratios, such as return on assets, or return on funds employed.

The last few years have seen a missionary zeal among those who would reform public sector accounting. Accrual accounting, it is claimed, will fix all the problems of valuation. A few deft keystrokes and computer based accounting systems will be able to produce bottom line performance indicators.

The process is not simple, however. Governments often face situations not shared by the private sector. How do we measure the value of pre-school education? What is the Sydney Opera House worth? Who owns the Hume Highway? There are practices which can be imported from the private sector, but there are also unique situations in government; after all, most activities are in government because private markets fail. In order to be able to choose appropriate private sector models, it is necessary to understand and deconstruct some of the main accounting concepts and conventions.

Precision Based on Convention and Assumption

Accounting appears to be a precise discipline, with figures in reports and budgets carried to many significant places. Fosters Brewing Group Ltd. in 1999-2000 had sales of $3 206 200 000. The value of assets of Australia Post at June 2000 was $3 037 500 000.

Numbers like this carry an air of authority and precision, masking what is in fact a very imprecise discipline, dependent on a number of conventions in valuation and realization. (Realization is defined below.) There are good arguments for adopting these conventions, but there are also good arguments for adopting alternative conventions. In many cases if we adopted different conventions we would get very different measures for asset valuation, profit, return on assets, and even for such apparently unambiguous concepts as cash holdings.

Most of the definitions of concepts and conventions in this chapter are taken from a standard text by Robert N Anthony, which millions of students have studied over the years.⁽¹⁾ I have not included all the standard accounting concepts and conventions, especially those which relate to balance sheets which are less important in government accounting. Definitions of conservatism and materiality are drawn from Horngren and Sundem, a more recent accounting text.⁽²⁾

Concepts and Conventions

Below are the main concepts and conventions of financial accounting. They are taken from private sector practice, but all are applicable, usually with modification, to government. Note that this chapter is concerned with financial accounting; the separate but related discipline of management accounting is covered in Parts 3 and 4.

Money Measurement Concept

Accounting records show only facts that can be expressed in monetary terms.

Many of the important aspects of an enterprise are very hard to measure at all, and even if they can be measured it is often hard to translate those measures to money terms. For example it is very hard to measure the reputation of a school. It may be possible to measure its academic achievement, but it is not easy to translate to money terms. In private enterprise, when a company changes hands, there is often a component called goodwill, which is the difference between the purchase price and the tangible value of the assets purchased. For example, a restaurant may have only a few physical assets, but will have an excellent reputation, and this is reflected in the selling price. Because public sector entities very rarely change hands, goodwill is very hard to measure or even to conceptualize in monetary terms. Even when there is privatization, such as with the sale of Australian Airlines, there is difficulty because privatization is often accompanied by changes in the regulatory environment.

The valuation of intangible assets, like goodwill, presents major problems. Banks are willing to lend against tangible assets like buildings, land and machinery, but are much more nervous about considering assets like intellectual property. When governments account to the public should they include goodwill? Perhaps they should err on the side of caution and not try to include it. But when there is a re-organization, involving closure of a school or hospital, or breaking up a functioning work team, should not governments be called to account for closing down a going concern worth much more than the land and buildings?

 

'Going Concern' and Cost Concepts

Financial accounting assumes that a business will continue indefinitely and is not about to be sold. Assets are normally entered on the accounting records at the price paid to acquire them.

This means, in fact, that accurate asset valuation is not very important; what is important is to have an unambiguous and objective base of valuation. The most objective source of values of assets is their purchase price. These assets will then be depreciated at standard rates. The alternative concept, not usually accepted in the preparation of financial statements, is that assets are recorded on an immediate liquidation value, that is, what we could get in an immediate sale. In essence that's an opportunity cost basis of valuation.

Neither concept is entirely satisfactory. In times of even moderate inflation the purchase price of assets rapidly becomes meaningless. Assets subject to rapid technological change can lose value rapidly. What is the value now of a computer for which you paid $5 000 five years ago?

In the public sector these problems are compounded because of the long life of assets and their often specialized functions. The cost concept yields absurd figures when we apply it to an asset like the Trans Australia Railroad, completed in 1917. So too does the immediate liquidation concept. Who wants a railroad? How do we establish a market price when there hasn't been a railroad sold in recent history? For a less commercial asset, like a collection of archives, the problem is even more marked. For buildings and land, market valuation is often dependent on zoning regulations. In Canberra the old hospital on Acton Peninsula, a peninsula jutting into Lake Burley Griffin, was closed in 1992. The site was zoned for use as non-commercial, and became the home of the Museum of Australia. It would be worth much more if it were zoned as commercial, for example, and those who follow the debates about betterment tax in the ACT will be familiar with the way in which governments can create or destroy land value through regulation.

The opportunity cost of assets is very hard to measure. Mikesell talks about the problems of valuing St Michael's Church, a 12th century church on the outskirts of London where a new airport was planned.⁽³⁾

In fact heritage assets present special problems. To the community they may be very valuable, to the government entity, responsible with a specific function, they may not be worth much at all. Sydney's General Post Office obviously had heritage value, but, being designed in a different era, it was not particularly suitable for mail handling. Lofty ceilings make heating and air conditioning expensive, confined access at the rear of the building makes it difficult to handle fork lift trucks. One possible approach to heritage assets is to use deprival value; that is, what would the entity lose if deprived of this asset? A heritage lighthouse, for example, could be replaced functionally by a light on a steel pole; the cost of such a hypothetical structure would represent the deprival value.

Another possible basis of valuation is to use the replacement cost concept. Usually after the lapse of any time it is unlikely that a similar asset would be installed in place of the existing one. A 2001 Sydney Harbor Bridge would be very different from the 1932 bridge; a single Canberra hospital would be constructed more centrally than Woden. It would be very costly for government agencies to obtain continuous updates on the cost of replacing their assets.

Accrual Concept

Income (or earning) is measured as the difference between revenue and expense rather than the difference between cash receipts and disbursements.

This leads to a consideration of the difference between the accrual concepts of revenue and expense and the cash concepts of receipts and disbursements. Revenue and expense are defined in accounting texts as transactions which change the owners' equity. A farmer sells 50 cattle off her farm for $300 each; these had been valued in the books at $20 each. Her expense in this transaction is $1 000 (50 * 20), her revenue is $15 000, and her income is $14 000. She does not write a check for $1 000, however. Neither does she necessarily get a check for $15 000; that may simply be credited to an account with a firm of stock and station agents, but it is still counted as revenue.

We can accrue expenses without any cash outlay. Our staff accrue long service leave entitlements, and that should be carried in our books as an expense and an unfunded liability. We can accrue revenue without any immediate cash receipt. The purchaser may take 30 days to pay, for example.

Another way to think of accrual is in terms of economists' concepts of value added as the difference between resources used and resources produced. It's an attempt to align accounting with economics.

The most contentious accrual concept is depreciation. When we buy an asset it is not immediately used up, but we draw down on its value over its lifetime. Depreciation is an accrued expense, but without any cash outlay. It is nearly impossible to determine how quickly we 'use' an asset; that is why we usually use standard depreciation rates. As accounting texts stress, depreciation is no more than a means of converting an expenditure on an asset into an expense; it is not a measure of wear or loss of utility. (Yet many people unthinkingly use financial accounting figures on depreciation as if they are "true and accurate" cost measures. We see this with motoring organizations which use standard depreciation rates as part of their estimates of the cost of operating a car. To more robust ways of measuring the cost of holding an asset over time, we return in Chapter 13.)

These difficulties in asset valuation, depreciation, and the cost of long tail liabilities all mean that accrual accounting is necessarily very imprecise, resting on arbitrary assumptions and standard methods of treating transactions, as laid out in accounting standards. (In government the key document in Australia is Australian Accounting Standard 29 Financial Reporting by Government Departments.) In financial accounting consistency is more important than accuracy. Unfortunately, many users of financial accounting information are unaware of this imprecision and the arbitrariness of the assumptions, and come to vest financial accounting figures with too much accuracy. Possibly this is a legacy of the tradition of cash accounting, for cash figures were objective and accurate - they were not particularly useful, however.

In spite of its limitations, the accrual concept aligns with common sense. When we "save" cash by putting a purchase on our credit card, we don't (or shouldn't) think of that as a saving; we have still incurred the expense. When we spend $1500 on a new refrigerator we don't consider that as immediate consumption, for we hope to get many years of service from that asset.

 

Realization Concept

Revenue is recognized at the time it is realized.

This sounds almost tautological, but it relates to the accrual concept, and the 'real' transaction that takes place, such as the delivery of the good or performance of the service, rather than the payment of the account. Similarly for expenses; they are recognized once they are incurred, not when we get around to paying the bill.

Another concept used particularly in the public sector is that of commitment. If, in January, we sign a contract to paint the building for $100 000, that is a commitment, but is not at that stage an expense. The painter completes the job in March. At that stage we incur an expense, which is recorded as a liability. We pay the painter in May, at which stage we incur a cash outlay, and erase the liability. The forward estimate process is an attempt to encapsulate commitments in projections, and to overcome the bureaucratic convention of wedging (incurring a small outlay now but with escalating commitments down the track). The commitment process can also be used to tie up funds as a deliberate means of limiting flexibility.

 

Conservatism Convention

The conservatism convention means selecting the method of measurement that yields the gloomiest immediate results.

Horngren and Sundem elaborate on this convention "Anticipate no gains, but provide for all possible losses, and if in doubt, write it off." Anthony refers to recording an event at a value such that the owners' equity is lower than it would be under any other basis of recording.

This can clash with the cost concept, and there are debates over which should be the overriding concept. Should we record our old computer at cost less depreciation, or should we write it off and record it as an extraordinary loss?

In the private sector, application of the conservatism convention can lead to undervaluation of corporate assets, thus leading to low share prices and the threat of takeover.

In the public sector the problems are more to do with accountability. Should the Municipality of Tinned Dog adopt the conservatism convention, reporting the airport at Tinned Dog to be worth only $100 000, which was the price of land and buildings in 1952? If so, is it misleading the community by understating the value of assets under its control and reporting a high rate of return? What is a reasonable benchmark for a return on the Tinned Dog airport? There is a good public policy argument that the airport should be valued at least at its opportunity cost.

 

Materiality Convention

An item is material if it is sufficiently large that its omission or mis-statement would tend to mislead the user of the financial statements under consideration.

Accountants often quote their favorite (possibly their only) Latin phrase: De minimus ne curat lax - "Of trivia the law cares not". This is not an excuse to dip into the petty cash, but it does justify a degree of what would otherwise be called 'misappropriation'. For example, under the accrual and realization conventions, if we buy $100 worth of postage stamps, these should be treated as inventory, and should become an expense only as we use them. Thus the purchase would see a credit to cash of $100, and a balancing debit to inventory of $100. Each time we use a stamp we should credit inventory with 45 cents, and debit postage with 45 cents.

In fact any sensible accounting system simply treats the whole purchase of stamps as an expense, with one pair of ledger entries.

In some regards the public sector probably applies tougher standards of materiality than the private sector. There is still the legacy of line item budgeting, for example. Many agencies apply the concept of "attractiveness and portability" to decide whether items should be entered on an asset registrar. A $50 coffee percolator may be "attractive and portable", a $500 collection of legal texts may fail on both tests.

While the public sector may have tight rules on materiality, it does not always have rigid controls on program expenditure; program misappropriation is still easy in many agencies, especially when it comes to administrative costs which are not appropriated by program.

 

Expenditure and Depreciation

From the time an asset goes into use, it is considered to be wearing out, or depreciating. Some assets, like land, are not considered to depreciate (when we consider soil erosion this is not always a reasonable assumption). There are closely related concepts like depletion (the use of a finite natural resource) and amortization (the write-down of an intangible asset like goodwill). Stock is not normally depreciated, but is retained as an asset at purchase price or manufacturing cost (an ambiguous concept as we shall see when we look at costing), but if it is damaged it can be written off - its value is converted to an expense. Similarly bad debts, subject to certain provisions, can be written off. Some outlays, like research and development, marketing, and staff training are really long term investments, and should, strictly, be capitalized and depreciated, rather than being claimed as expense. By convention and regulation, however, we usually write off these as they occur, in accordance with the conservatism convention.

Because depreciation, write-offs, and other measures to convert assets into expenses increase reported expense, they lower the reported income of the entity. They do not change the cash position, however. There is, therefore, an incentive to use them to their full extent in the private sector, as corporate tax is paid on the reported income. Offsetting this incentive, however, is the fact that they reduce the asset base and the reported income, making the company look less attractive to its shareholders and creditors. There are many taxation and accounting regulations about depreciation and similar measures.

There are many conventions in depreciation, but the most common in use are straight line (also called prime cost) and diminishing value (also called declining balance). Under the straight line method, the value of an asset is written off in equal increments each year; under the diminishing value method an equal percentage of the value of the asset is written off each year. If a new asset is used only part of the year, the depreciation claimable relates only to the months it was in use; thus for an asset installed in April only 2/12 of a year's depreciation can be claimed. The tables below compare straight line and declining balance depreciation methods for an asset with a purchase price of $4000. (It is a convention that a higher initial rate of depreciation is used in declining balance depreciation than in straight line depreciation.)

When a depreciated asset is sold, any receipt above its residual written-down book value is a revenue, any shortfall is an expense.

Exercises

Assume we're in a government agency which requires us to depreciate at standard rates any assets costing over $1 000.

(1) In July 2000 we buy a computer for $5 000. We are allowed to depreciate it at 20 percent straight line. What depreciation do we claim as an expense in 2000-01, and what is its book value on 30 June 2001?

(2) For the same computer, what depreciation can we claim in 2001-02, and what is its book value on 30 June 2002?

(3) We sell the computer in January 2003. How much depreciation can we claim over 2002-03. What is its book value when we sell it?

(4) We receive $2 400 for the computer when we sell it. How much of this is income?

(5) In December 2000 we buy an ambulance for $100 000. We are allowed to depreciate it at 15 percent straight line or 22.5 percent diminishing value. How much depreciation can we claim in 2000-01, and what is its book value on 30 June 2001?

(6) For the same ambulance, what depreciation can we claim in 2001-02, and what is its book value on 30 June 2002 (You will need a calculator.)

(7) We have a compactus filing system which we bought many years ago for $25 000. It can be depreciated at 10 percent straight line. Its book value on July 1 2000 was $625. How much depreciation can we claim in 2000-01?

A completed spreadsheet, showing answers to this exercise, is at ch04ex01.xls

Accrual and Cash Accounting in the Public Sector

Traditionally central government agencies have used cash based accounting, while government business undertakings have been using accrual systems for many eyars. Entities permitted to operate with trust funds are generally required to use accrual accounting. Unfortunately the debate is becoming something of an 'either/or', without the acceptance that even private entities which keep accounts on an accrual basis also prepare cash flow statements reconciling cash transactions with the accrual based income statement.

While central budgets and cash appropriations remain the main instruments of central control, cash accounting will retain a high level of importance in the public sector. At the expense of some simplification it's a question of primacy. In my view, in the private sector the cash flow statement is subsidiary to the income statement (also called the profit and loss statement); in the public sector the cash flow statement will remain the primary level of report and control, while an accrual based income statement will be secondary.

Exercise

To see how such a pair of statements might work, let's take a set of transactions over a year in a small government office:

(1) Receives two appropriations, of $120 000 in July, of $200 000 in October.

(2) Pays salaries of $200 000 and pays superannuation on these payments at 10 percent of salaries.

(3) Pays rent in four transactions in advance:

• September - $8 000 for period October-December
• December - $8 000 for January-March
• March - $8 000 for April-June
• June - $9 000 for July-September

The bill for July to September was $8 000, paid in June in the previous financial year.

(4) Early in January buys a new car for $16 000. This will be depreciated at 15 percent a year on a straight line basis. The old car was traded in for $3000; its book value at July 1 had been $2600, and over the half year it accumulated depreciation of $1650.

(5) Does a consultancy contract for a customer, valued at $8 000. This was completed in June, but the bill was not paid as at the end of June..

(6) Pays four phone bills, covering the preceding quarter:

• July$2 500
• October$3 200
• January$2 800
• April$3 100

The bill which arrived in July was for $3 200.

(7) Pays $43 000 in other minor expenses, all accruing in the month immediately preceding bills. This included $4 000 of previous year's accounts paid in July. Bills accrued in June but which were not paid amounted to $6 000.

The statement on a cash basis, is at ch04ex02.xls. The first sheet has the cash flow statement only with space for an accrual statement; the second sheet is completed with notes.

Exercise

A hospital pharmacy has an annual drug budget of $1.2 million. It is early June and only $20 000 remains unspent.

The pharmacy is well stocked except for drugs for acute antihypertensive treatment, and must have at least 20 doses on hand - 15 representing normal monthly use, 5 being an emergency stock. There is no option but to keep the emergency stock.

The drug company will supply 100 doses at $700 each, cash on delivery. It will not supply smaller orders, however. A large wholesaler will supply 100 doses at $750 each, with thirty days to pay. A small wholesaler will provide small quantities (ten or more) at $1000 each dose, cash on delivery.

What should the pharmacist do? How might he or she justify the decision to a Parliamentary committee?

Balance Sheets and Financial Ratios

Those who have studied accounting will realize that of the three key business statements, the income statement, the cash flow statement and the balance sheet, we have covered only the first two.

A balance sheet is a snapshot of the financial status of an entity, usually prepared at the end of each accounting period. It is presented in the following form:

The balance sheet, by definition, balances. Therefore the total assets = total liabilities and equity. Manipulating this equation gives us:

owners' equity = total assets - total liabilities

In other words, the owners' equity is a residual item, colloquially the value left over after covering debt.

In the private sector the return on owners' equity is a prime measure of corporate health. For example, we have Foster's balance sheet as at 30 June 2000:

Note the owners' equity is $2 307 million, which is the balancing item after $2 794 million of liabilities is offset against $5 101 million of assets . Further on in the report we see from their profit and loss statement that their profit before tax and interest was $651 million, giving a return on assets (AKA return on investment or ROI and less commonly return on funds employed or ROFE) of 12.8% (651/5101). Interest payments were $65 million, leaving $586 million of profit for equity holders. This gives a return on equity (ROE) of 25.4 percent (651/2307). Tax payable was $155 million, leaving an after-tax return for equity holders of $431 million, or a return on equity after tax of 18.7 percent (431/2307). These figures are high by industry standards - ROEs after tax are usually in the range of 10 to 12 percent.

At ch04ex03.xls is a spreadsheet in which you can calculate and confirm these values. The completed sheet also allows you to change the value of non-current assets, and to see how that valuation changes the financial ratios, particularly the rates of return - ROI and ROE.

These figures, of course, are based on Fosters' financial statements. If their financial asset valuation were true and accurate, we would reasonably expect their share price to be equal to the amount of owners' equity divided by the number of shares on issue. Fosters has 1 726 million shares on issue, implying a share valuation of $1.34. (2 307/1 726). In fact, Fosters' share valuation over 2000 was consistently between $4.00 and $4.60. Share values are based on investors' expectations of future earnings - the net present value of those earnings. (We will cover the concept of present value in Chapter 9). Another way of looking at this is that they are valuing their assets at a much higher level than their book value. A share valuation of $4.00 implies a market value of owners' equity of $6 900 million (1726*4), or assets of around $9 700 million, rather than $5 100 million. That would imply a ROE of around 8.4 percent.

This example is typical of private companies; markets generally value the assets of firms at a much higher level than book value. Many old assets are depreciated to zero, but are still providing good service. Many real estate assets, such as hotels, are probably under valued. Many assets ("things of value") are not recorded as "assets" at all - reputation in the market, the skills of brewers, the value of the VB brand. Indeed, over time, as human capital becomes more valuable and as physical capital becomes less expensive, the value of assets shown on financial statements is coming to be less and less a reflection of the actual value of a firm's assets.

In the public sector, however, there is often a dogmatism about the "truth and accuracy" of financial statements - reflected in publications such as the Commonwealth's financial management work Beyond Beancounting.⁽⁴⁾ As well as problems in asset valuation, figures such as profit are manipulable; profit can be made to look good by cutting down on training, maintenance and other "expenses". And, as we have suggested, asset valuation in the public sector is much more problematic than it is in the private sector, and the meaning of performance is ambiguous. For example, does a high profit in a GBE reflect efficient management or exploitation of a monopoly position - a point to which we return in Chapter 15.

Notes

General References

R G Walker "Accrual Accounting - Necessary but not Sufficient" Ch 15 in Guthrie, J; Parker, L; Shand, D The Public Sector - Contemporary Readings in Accounting and Auditing (Harcourt Brace Jovanovich NSW 1990)

Specific References

1. Robert N Anthony Essentials of Accounting (Addison Wesley 1964)

2. Charles Horngren and Gary Sundem Introduction to Management Accounting (Prentice Hall 1990)

3. John L Mikesell Fiscal Administration - Analysis and Applications for the Public Sector (Brooks/Cole CA 1991)

4. MAB/MMIAC Beyond Beancounting - Effective Financial Management in the ABS - 1998 and beyond (AGPS 1997).