Back Matter
Author:
Alberto Soler 0000000404811396 https://isni.org/isni/0000000404811396 International Monetary Fund

Search for other papers by Alberto Soler in
Current site
Google Scholar
PubMed
Close
and
Mouhamadou Sy 0000000404811396 https://isni.org/isni/0000000404811396 International Monetary Fund

Search for other papers by Mouhamadou Sy in
Current site
Google Scholar
PubMed
Close

Annex 1. Structure of the Template: Inputs and Outputs

Input 1–Basic

The following data are entered: main country and sector of operation of the SOE, and last year of observed financial data. By default, it is assumed that the template generates projections from the next year onwards. Verify that formulas are in place for all projection years across the template.

Input 2–Income Statement

The user enters available observations from the Income Statement for at least the last two years.

Input 3–Balance Sheet

The user enters available observations from the Balance Sheet Statement for at least the last two years.

Input 4–Assumptions

This worksheet, referred to the baseline scenario, calculates by default macroeconomic projections drawing on the IMF’s World Economic Outlook (WEO). The user can replace these formulas by projections from other sources. Interest rates of assets and liabilities are not provided and must be entered by the user. It should also verify that the WEO provides projections for all relevant years. But the user can use her own macroeconomic projections

Based on the assumptions entered by the user, the worksheet automatically calculates the growth in the volume of sales in domestic and foreign prices. These variables are used in the Income Statement to project gross sales. The user can also rely entirely on projections of sales in volume and prices from external sources, such as the SOE’s management, by replacing the existing formulas by those projections. Additionally, the share of domestic to total sales in local currency (i.e. the sales tax base) is estimated in the same worksheet and applied in the Income Statement. It will be important to ensure the projections are credible and robust (for example including a critical assessment of the projections prepared by the SOE or other external sources).

The assumptions relative to the structure of debt allow for a relatively flexible modeling of new borrowings. For instance, the structure of both loans and debt securities can be replicated by combining different grace periods and maturities of new borrowings (for example grace period plus one-year maturity for debt securities). Grace periods and maturities can be different for borrowing denominated in local and foreign currency. The user can also set different shares of government’s loans and government guaranteed debt to total debt for pre-existing debt before the start of projections and new borrowings.

Investment assumptions in the baseline scenario may reflect the original business plans of the company but can also be subject to sensitivity exercises to analyze the effects of alternative investment paths. This possibility becomes more relevant when both SOE and government face severe liquidity constraints, and neither borrowing nor capital injections are a viable alternative. On the other hand, purchases of fixed assets can reflect one-of operations, or the construction of assets over several periods as part of turnkey projects or other PPP contracts.

When entering the assumptions, it is important to ensure they are consistent. For example, a heavy expansion of the SOE’s productive capacity can translate into a higher elasticity of sales to GDP than observed in the past. Likewise, exchange rate movements should be key drivers of revaluation effects if a high share of the company’s assets is denominated in foreign currency. The assumptions may also be influenced by the legal and regulatory transparency of the environment where they operate (for example an ineffective legal system can entail a higher level of delinquent receivables).

Benchmarking

The sheet automatically compares the SOE to SOEs in the same sector regarding profitability, liquidity, leverage, and efficiency indicators. The comparison is done with the whole sample of countries, and with the sub-set of countries of a similar income country group if data are available.

Stress Test

This worksheet plays an analogous role to Inputs 1–4 but refers to the stress scenario.

The user is asked to design the stress scenario by entering deviations from the baseline of macroeconomic and market-specific variables, by year. These deviations are additive, except for the shock on the level of receivables and their share to materialize, which are multiplicative (that is, the values of expected recovery of receivables in the stress scenario are obtained by multiplying their pre-shock values by these coefficients).

Note that, when defining the size of the shocks, the user can take into consideration both first and second-round effects. This is the typical case of variable interest rates on the company’s debt. The immediate impact of a growth shock may be to raise the risk premium of the company if its revenues drop. However, if the shock also entails an exchange rate depreciation, and a significant share denominated in domestic currency is held by foreign investors, an additional increase in rates could be expected. Thus, the user may want to reflect both effects in the deviation of stressed domestic rates from the baseline.

Parametric assumptions and business plans are, by default, the same as in the baseline scenario. However, the user can modify them in the stress scenario if needed, just by replacing the formula in these cells by different numerical values. This degree of flexibility is intended to increase the realism of the scenarios, as behavioral parameters are often asymmetric throughout the baseline cycle. For instance, a high utilization of the productive capacity of the company may imply a low elasticity of sales to real GDP growth, but this elasticity may be considerably higher during an economic downturn, or capital gains should be commensurate to receipts from asset sales. Analogously to Assumptions, this worksheet calculates sales volume growth, domestic and foreign prices developments and the share of domestic to total sales in the stress scenario.

Debt Projections

This worksheet, referred to the baseline scenario, generates debt projections by importing debt data and their related parameters from Input 2, 3 and Input 4-Assumptions.

Outstanding stocks by type of debt are calculated as previous stocks, plus borrowing less amortizations.

Borrowing is imported from the Cash Flow Statement and apportioned between FX and local-currency denominated debt according to parametric assumptions. These also determine the amount of government-guaranteed debt and government loans out of this borrowing. Borrowing in foreign currency is obtained as a share of total borrowing in local currency, and then converted into foreign currency at the period’s exchange rates.

Amortizations of debt outstanding at the beginning of the projection period and debt issued after that year are calculated based on the assumptions about the average maturity of debt and, for new borrowing, grace periods. For the first year of projections, amortizations are given by observed data on current liabilities in Input 3. Since uniform assumptions are applied for the amortization of all local currency- and FX-denominated debt, when debt instruments within each category are very heterogeneous this can lead to some loss of accuracy in the amortization profile. In these cases, the user may wish to replace the formulas that compute the amortization profile of pre-existing debt by the sum of expected amortizations for each instrument.

Interest payments are also calculated in this worksheet. This is done by applying an effective interest rate to the stock of debt outstanding by the end of the last year. The effective interest rate is projected as a linear combination of last year’s effective interest rate and the interest rate of marginal borrowing entered in Assumptions. The parameter that captures the variability of interest rates is entered in Assumptions, and it should be understood as a weighted average for the whole stock of debt.

The amortizations and interest payments of government-guaranteed debt and government loans are derived by applying their share in total domestic debt (entered in Assumptions) to domestic debt amortizations and interest payments on domestic debt.

Amortizations and interest payments of foreign-currency debt are calculated in their currency of denomination and converted into local currency at the average exchange rate entered in Assumptions. The value in local currency of the outstanding stock of foreign debt is obtained by applying the end-of-the year exchange rate entered in Assumptions.

Debt Projections (Stress)

Analogous worksheet to baseline Debt projections but refers to the stress scenario.

Income Statement (Baseline and Stress)

These worksheets import all necessary data from other worksheets of the template and automatically calculate the main components of net income (pre- and after- income taxes). It also calculates the dividends distributed to the government, based on after-tax profits. Line items are projected by making their last observed values grow according to the rates given by the parameters in Input_5 Assumptions and Stress Test.

Among operating revenues, gross sales are determined by combining the growth rate of their volume and prices in domestic and foreign markets, expressed in local currency. Net sales are projected by subtracting sale taxes from gross revenues. Other revenues are obtained by indexation to domestic inflation but could be augmented by the growth rate of gross sales if the turnover of the primary and secondary activities of the company are correlated. Subsidies are entered exogenously by the user in Input 5 Assumptions and Stress Test.

Among operating expenses, personnel ones grow along inflation (adjusted by a degree of indexation entered by the user) and employment plans, and costs of goods and services depend on the volume of inputs purchased (linked, in turn, to the volume of production) and their prices in domestic and foreign markets. Other operating expenses are indexed to inflation, but their formula can be adjusted to add some contract-specific items, such as availability payments in PPPs.

Non-operating revenues are given by: i) interest receipts are obtained by applying short and long-term interest rates (domestic and foreign) to asset holdings; ii) dividends are indexed to nominal GDP growth; iii) capital transfers are determined in the Cash Flow Statements; iv) other revenues have two components: a first one is linked to inflation, and a second one to realized capital gains, as entered exogenously by the user.

Non-operating expenses mainly comprise interest payments (imported from Input 4_Debt and Debt Stress) and other expenses, which have a symmetric structure to other non-operating revenues. Other non-operating expenses non linked to inflation also comprise receivable write-downs, and they could be augmented by an additional row that captures the realization of contingent liabilities (legal claims, termination payments in PPP contracts, etc.).

The CIT paid is obtained by applying the exogenous rate to the pre-tax profit, once subsidies and capital transfers are excluded -they are assumed to be exempted from the tax base, but the user can easily modify the formula if this does not hold for a specific firm-.

Sometimes the user may identify outliers (i.e. exceptionally high or low values) in the last observations of some line items in the Income Statement. This can only be done if the time series of past observations includes at least three or four years. In these cases, it is important to adjust the projection of the concerned line item to avoid carrying the anomaly into the whole projection period. This adjustment can be made, for instance, by linking first year’s projection to the data which precedes the outlier rather than to the outlier, or to an average of values in the last 3–4 years.

Cash Flow Statement (Baseline and Stress)

These worksheets also operate automatically. They comprise three blocks.

  • Calculation of end-year cash balances. This is done by aggregating operating, investment and financial cash-flows, plus capital injections, to end of last year’s cash balances. Deriving these projections does not require making any assumption on the growth rates of the variables but linking them to the relevant line items of the Income Statement/Debt worksheets and changes in the Balance sheet stocks. Operating cash flows are obtained by adjusting operating revenues and costs by receivables, payables, changes in inventories, depreciation, and income taxes. Investment cash flows reflect net transactions in property, plant and equipment and investment property. Financial cash-flows stem from differences in the stock of debt, as well as net interest and dividend payments.

  • Calculation of borrowing and capital injections. First, the template estimates equity and cash balances before new borrowing and capital injections and computes the liquidity gap as the distance to the liquidity target. Overdrafts are ruled out by formula. Then, the tool determines the level of cash balances necessary to reach the targeted quick ratio, and the liquidity gap as the difference between this target and the cash balances before borrowing and capital injections. Subsequently, the tool quantifies the share of borrowing and capital injections to fill the liquidity gap and, simultaneously, complying with the leverage cap.1 The liquidity gap will be first filled with borrowing, and only residually with capital injections. The tool also estimates the amount of these injections which help the company to address its guaranteed debt service, under the assumption that these needs are the first to be met by means of capital injections.

  • Estimation of the counterfactual scenario without capital injections, in which liquidity gap is entirely filled through borrowing. All additional borrowing in the counterfactual is assumed to be domestic, its grace period is one year for simplicity and its maturity the same as when capital injections are present. The counterfactual also includes interest payments for additional borrowing, by multiplying the same effective interest rate applicable when capital injections are made to the additional borrowing. Liquidity needs are computed by adjusting the value obtained with capital injections by the higher current long-term liabilities without injections, annual additional amortization, and incremental interest payments.2

Balance Sheet (Baseline and Stress)

Essentially automatic worksheets. They include two blocks: (1) balance sheet in simplified IFRS terms,3 where projections are made; and (2) mapping of IFRS balance sheet into GFSM 2014.

Balance sheet items with a high turnover are linked to the Income Statement. This is the case or receivables and inventories (linked by means of stable ratios to gross sales) or payables (linked through an analogous ratio to the cost of goods and services). Defined benefit liabilities are linked to the wage bill, as this tends to be the basis for their calculation.

End of period cash balances are imported from the Cash Flows and Cash Flows Stress.

Long-term liabilities and current long-term liabilities are imported from the Input 4_Debt, the latter being given by next period’s amortizations. When amortizing debt is denominated in foreign currency, they are valued in local currency at end-of-period exchange rate.

Fixed and financial assets follow their own accumulation equations. According to them, the value of the stock at the end of every period is equal to the value at the end of the previous year (less depreciation, for fixed assets), plus net acquisitions in each period. Net acquisitions by year are entered by the user in each period in Assumptions.

Some assets (receivables, fixed and financial assets) are subject to revaluation effects in each period, their amount being entered in Assumptions and Stress Tests. These three types present specificities regarding the application of revaluation:

  • Receivables. A stock of doubtful receivables is automatically estimated every period, and the value of receivables calculated by formula is adjusted by changes in this stock. This is done under the assumption that there is a very high probability that doubtful receivables do not generate any cash flow for the company in the next period. In the baseline scenario, the value of this stock is entered in Assumptions under “provision for doubtful receivables,” while in the stress scenario they are generated by the shock on the expected share of receivables to materialize. When the value of this parameter is lower than one, it denotes an increasing balance in the stock, whereas vales higher than one imply decreasing doubtful receivables.4

  • Fixed assets. The revaluation effect adds to the value of the undepreciated stock at the end of the previous period and is part of the value at the end of the current period.

  • Financial assets. The treatment is similar to fixed assets, although depreciation is zero for these assets. Note that revaluations, as entered by the user in Assumptions, should take into consideration both developments in assets denominated both in local and foreign currency, as well as exchange rate movements.

Other current and non-current asset and liabilities are held constant at the level of the last observed value, given the heterogeneity of these items.

Shareholder’s equity encompasses three main elements:

  • Shareholder’s capital fed by capital injections.

  • Accumulated gains, the net inflows of which are imported from the Income Statement.

  • Accumulated Comprehensive Income. In a simplified format, this item captures unrealized valuation effects of assets and liabilities including those caused by exchange rate changes.

The mapping of the IFRS BS into the GFS assumes that equity is calculated residually as the difference of the asset minus the liabilities of the company. This is possible if the equity is fully owned by the government, or if its shares partially owned by other units are not traded. Other simplifying assumptions are also made, such as the valuation of assets at market prices on books or the equality between the face and the market vale of debt. Other non-current assets in IFRS may comprise both financial and non-financial assets in GFS terms, and their apportionment is made by a means of a coefficient entered in Assumptions. The SOEs assets and liabilities, once mapped to GFS, are consolidated across public sector units by means of coefficients entered in Assumptions, that are held constant over the projection horizon. Regarding equity, the user enters the initial share owned by the government as an assumption, and the tool recalculates it every period taking into consideration capital injections.

GFS (Baseline and Stress)

Fully automatic worksheet. It is structured into three blocks.

  • Calculation of borrowing needs. Revenues and expenses are expressed in accrual terms and linked to the relevant items of the Income Statement. Net transactions of non-financial assets are taken from netting out acquisitions and divestment in these assets as entered in Assumptions and Stress Test, and assuming that these operations are always fully paid in the same year they are accrued.

  • The second block depicts net transactions in financial assets and liabilities, their difference matching lending capacity/borrowing needs. These transactions reflect changes in stocks, as projected by the Balance Sheet. Unrealized valuation effects and asset write-downs are removed above and below the line, as these would be part of the Statement of Other Changes in Assets and Liabilities in GFS.

  • Derivation of gross financing needs in the absence of capital injections.

Output 1–Performance (Baseline and Stress)

These worksheets summarize, the main financial and economic indicators of the company, regarding profit-ability, liquidity, solvency, efficiency, and its financial position. Financial indicators are summarized in Table A.1 below. On top of them, the worksheet also calculates some ratios indicative of the company’s productivity and efficiency (operating revenue per employee, labor cost per operating revenue and average personnel cost per employee), and the SOE size (number of employees, total assets in local currency, and assets and liabilities to GDP).

Most commonly accepted high-risk thresholds are 0 for profitability ratios; 1.25, 0.8 and 1.2 for the current, quick and interest coverage ratios respectively; 1.5, 0.75 and 0.5 for debt-to-equity, debt-to-assets and non-current liabilities to assets ratio respectively.

Most financial indicators are projected in each scenario in two situations: after capital injections and subsidies, and in a counterfactual where the company bailouts are replaced by additional borrowing (see description of Cash Flows and Cash Flows Stress).as Flows et above).

Output 2–Charts (Baseline and Stress)

In this case, the summary of results is done through charts, which can be selected by means of dropdowns menus.

The charts are imported from two worksheets named RAW CHARTS and RAW CHARTS (ST). In these worksheets, the user will find the data associated to each chart and will be able to modify their content or format.

Annex Table 1.1.

Financial Ratios Projected by the Template

article image

Output 3–Relations with the Government (Baseline and Stress)

This is a breakdown of budgetary and balance sheet ties between the government and the SOE.

On the one hand, these worksheets calculate the net inflows into the budget in accrual for each projection year, in both scenarios. These inflows are derived as the sum of taxes, dividends and interest payments, less subsidies and transfers. Taxes are drawn from the Income Statement, dividends computed as the product of the net income after taxes by the dividend pay-out ratio entered in Assumptions. Subsidies are exogenously entered in Assumptions, and capital transfers determined as the share of capital injections non-commercially remunerated. The NPV of these flows is also estimated, as a percent of the pre-shock GDP. Discount rates draw from the baseline long-term interest rate projections, but they can be modified if they are not deemed to be risk-free discount rates.

In addition, the worksheet displays the stock of outstanding government loans to the SOE, its equity and guaranteed debt. The first two items are assets for the government and consolidated in the public sector-wide balance sheet, and the latter is a contingent liability for the general government.

References

  • Allen R. and M. Alves. 2016. “How to Improve the Financial Oversight of Public Corporations.” IMF How to Note 16/05, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Baum, A., P. Medas, A. Soler, and M. Sy. 2020. “Managing Fiscal Risks from State-Owned EnterprisesIMF Working Paper 20/213, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • Baum, A., C. Hackney, P. Medas, and M. Sy. 2019. “Governance and SOEs: How Costly Is Corruption?IMF Working Paper 19/253, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • FAD. 2008. Unpublished. “Indonesia: Assessing Fiscal Risk from State-Owned Enterprises. Technical Assistance Report.”

  • International Monetary Fund (IMF). 2005. “Public Investment and Fiscal Policy—Lessons from Pilot Case Studies.” IMF Policy Paper, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • IMF. 2007. “Public Enterprises and Fiscal Risk—Lessons from the Pilot II Country Studies.” IMF Policy Paper, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • IMF. 2012. “Fiscal Transparency, Accountability, and Risk.” IMF Policy Paper. International Monetary Fund, Washington, DC.

  • IMF. 2016. “Analyzing and Managing Fiscal Risks: Best Practices.” IMF Policy Paper, International Monetary Fund, Washington, DC.

  • IMF. 2018. “Fiscal Stress Tests for The Gambia.” Selected Issues Paper, IMF Country Report 18/100, International Monetary Fund, Washington, DC.

    • Search Google Scholar
    • Export Citation
  • IMF. 2020. “State-Owned Enterprises: The Other Government” in Fiscal Monitor. Washington, DC, April.

1

The relationship between budget accounts and SOEs depends on the type of coverage. For this analysis, fiscal risks and costs (when risks materialize) refer to the risks to the level of government that owns the SOE.

2

The contribution of SOEs to government revenues is significant in some countries, particularly in oil exporters.

3

The expectation of a bail-out may be particularly strong if the SOE has limited access to capital markets or if it has strategic importance for the government, such as providing core services (for example, water or electricity), and faces financial distress.

4

Contingent liabilities are not part of the government balance sheet, but they should be appropriately disclosed (see IMF’s Fiscal Transparency Code).

5

See FAD (2008): Indonesia: Assessing Fiscal Risk from State-Owned Enterprises and IMF (2018): Fiscal Stress Tests for The Gambia.

6

The template does not explicitly model PPP or PPA operations. Nonetheless, some elements of these operations can be easily embedded in the formulas if needed (see Annex 1).

7

The template uses the IMF’s World Economic Outlook projections, but users can enter their own projections.

8

Market-specific factors can also significantly influence sales beyond broader macroeconomic dynamics. For instance, this may happen due to product-cycles, whether the SOE faces increased competition in the market, or a key client undergoes financial difficulties. The user may optionally enter these projections when relevant.

9

Optional parameters and projections require a good understanding of the dynamics of the market where the SOE operates, as well as its portfolios of assets and clients. These features of the tool can be phased-in and be brought into a more advanced level of analysis.

10

A true measure of efficiency should consider all the inputs of the firm. However, revenue per employee is a useful proxy for efficiency, given data constraints, allowing for comparison across SOEs.

11

See Baum, Hackney, Medas, and Sy (2019) for the treatment of the data. For the benchmarking exercise, SOEs are defined as commercial organizations that are ultimately owned by public sector entities. The ownership condition requires that central or subnational governments own or control more than 50 percent of the firm.

12

There are other possible rules to determine capital injections but combining liquidity and solvency indicators strikes a balance between mitigating the risks of default, and moral hazard considerations. Moreover, since the user can flexibly set the liquidity and leverage thresholds, this allows factoring in different government’s preferences and strategies, as well as the fiscal space available for conducting bailouts.

13

Receivables are presented net of write-downs for prudential reasons, as it assumed that there is a very high probability that those receivables considered as doubtful will never generate cash flows for the SOE.

14

The share of assets and liabilities to consolidate in the broader public sector balance sheet (and calculate impact on net worth) can be entered by the user as financial assumptions.

15

Royalties should only be considered for commodity-producing companies, particularly oil.

16

See Annex 1 for more details about the derivation of the effective interest rate.

17

Liquidity and leverage thresholds of the tool are used only to operationalize the linkages between the SOE’s performance and government’s accounts. In practice, bail-out decisions are entirely discretionary. Moreover, lack of governmental willingness to assume the cost of implicit contingent liabilities can be signaled by setting very low liquidity floors or very high leverage thresholds.

18

For example, if the governments do not expect to be repaid, or the rate of return is lower than for other investments comparable in terms of risks and maturity—or is lower than the cost of borrowing by the government.

19

As for early warning values, the debt-equity ratio varies across industries, but usually values above 1.5–2 denote high risks. The preferred quick ratio is above 1, and values below 0.5 tend to denote high risk. The quick liquidity ratio is given by cash and deposits, plus receivables and other current liquid financial assets, divided by current liabilities. The debt-to-equity ratio equals total liabilities divided by shareholder’s equity. This double threshold seeks to strike a balance between short-term risks (liquidity shortages) and medium-term risks (solvency).

20

Note that the liquidity gap implies constraining the way gross financing needs are met. In general, gross financing needs can be met either through a net reduction of assets or the incurrence of liabilities. Another option to fll this gap is slowing down net purchases of assets, including through divestment. The template also allows a resort to this option, as net investment plans in financial and non-financial assets are entered exogenously by users.

21

2018 profits were exceptionally high because of the writing-of of some provisions.

22

Since the share of receivables to gross sales is likely to be pro-cyclical, the impact of the growth shock on the cash-fow will often be higher than on operating income (which is reported on an accrual basis), unless payables experience a similar increase.

23

The tool offers the possibility of assessing risks for oil companies and can be easily expanded to other commodities. See Annex 1 for more details.

24

When the share of receivables at high risk is higher than zero, the SOE is assumed to write them of. When this share is negative, some of the receivables classified in previous periods as doubtful are recovered, and this gives rise to a decrease in the stock of doubtful receivables, and positive non-operating revenue.

25

The liquidity and leverage thresholds are maintained at 0.6 and 1.5, respectively, as in the baseline scenario.

1

The necessary level of cash balances to meet the liquidity threshold, defined in terms of the quick ratio, may be negative. This may happen if receivables are high by comparison to current liabilities, and/or the targeted ratio is low. This situation is ruled out by the template though, as it would imply an overdraft. In these cases, a minimum level of positive balances is required, and this will imply that final liquidity is above its threshold.

2

This counterfactual should be understood as a mere approximation to a full-fledged scenario and does not compute, for example, the effects of possible higher spreads on the SOE debt, or the shortening in the average life of debt as a result of worsening financing conditions.

3

If the probability of receivables becoming delinquent is deemed to be high enough, the user by write them of just by registering the associated losses in “other non-operating expenses” in the Profit and Loss Statement, and cancelling them on the Asset side of the Balance Sheet. Care should be taken that these write-offs are not computed as an increase in cash in the Cash Flow Statement.

4

If the user estimates that a share of receivables is only at moderately high risk and writing-of some of them may be premature, it can proceed in the following way: i) use the face value of receivables, without adjusting them by the value of the provision; ii) include the provision in the line “Other current liabilities” in the Balance Sheet; iii) make sure that the changes in the value of the provision are not considered in the cash flow statement.

  • Collapse
  • Expand