Expenditure Aggregates and Components of GDP at Constant Prices

1. Expenditure Aggregates and Components of GDP at Constant Prices UN-ESCWA 22 – 25 September 2007 Cairo

2. Main Expenditure Aggregates • Household Final Consumption Expenditure (HFCE) • Government Final Consumption Expenditure (GFCE) • Final Consumption of NPISHs • Gross Savings (B.8) – balancing item [discussed later, with Measures of Real Income] • Gross Fixed Capital Formation or GFCF (P.51) • Change in Inventories (P.52)

3. HFCE at Constant Prices – Issues of Compilation (1) • CPI (Laspeyres formula): The only source of comprehensive data on changes in purchasers' prices on consumer goods and services. • But, the ideal national accounts deflator: CPI - Paasche formula. • Under unstable conditions like hyper inflation, the difference between a Laspeyres formula CPI and a Paasche formula CPI can be significant. • Usually, base periods for CPI and national accounts constant price estimates differ.

4. HFCE at Constant Prices – Issues of Compilation (2) • Coverage of HFCE differs from the consumption basket used for CPI often. • Thus, simply deflating total HFCF by the total CPI is not enough. • It is necessary to prepare the HFCE estimates at constant prices at the most disaggregated level possible, using • detailed components of CPI as deflators and • additional price information for flows not properly covered by CPI, when appropriate.

5. HFCE and CPI - Difference of Coverage (1) Consumption of own-produced goods and services: These items are often not included in CPI (when based on 'final monetary consumption expenditure') • In NAs, ‘output for own final use’ is valued at basic price, or at costs of production, if market prices are not available. • Where output for own final use is a significant part of total consumption of a certain product, separately deflate • the part retained by the producer by a suitable PPI, and • the part marketed by the corresponding component of CPI.

6. HFCE and CPI - Difference of Coverage (2) Goods and services received as income in kind: Often excluded from coverage of CPI. • In NAs, these are valued at basic prices, if produced by the employer, and at market prices if purchased by the employer. • If the payment in kind to employees from own-production forms a significant part of household consumption for a product, it should be deflated using suitable PPI.

7. HFCE and CPI - Difference of Coverage (3) • Goods and services purchased abroad by resident households: CPI covers all purchases made by resident and non-resident households on the economic territory of a country. • In the national accounts, HFCE includes purchases made abroad by resident households and excludes the expenditure of non-residents in the economic territory. • If purchases abroad by resident households form a significant proportion of HFCE, then this part should be deflated separately.

8. HFCE and CPI - Difference of Coverage (4) • Pension funding and life insurance services: Life insurance services and pension services are usually excluded from the CPI, since implicit service charge is not directly observable. • These components of HFCE at current prices should be deflated by GDP deflator, as done for the production-side estimates. • Miscellaneous items such as the service charges for lotteries and gambling, prostitution and narcotics: These are often excluded from CPI owing to the practical difficulties. Use the same indicators as done for production-side estimates.

9. HFCE and CPI - Conceptual Difference (1) • Services of owner-occupied dwellings: These represent a high proportion of final consumption of households. • The recommended method: Deflating by a suitable index of actual rent levels. • Since no distribution margin involved, this component of HFCE at current prices should be deflated by the same indicators as done for production- side estimates. [See Handout]

10. HFCE and CPI - Conceptual Difference (2) • Non-life insurance services: ILO Manual on CPI recommends exclusion of this item from CPI basket, since implicit service charge cannot be observed. • EUROSTAT Handbook recommends not to use changes in gross premiums rates for deflation. • This component of HFCE at current prices should be deflated by GDP deflator, as done for the production-side estimates.

11. HFCE at Constant Prices - A, B and C methods • ‘A’ methods: deflation of HFCE at current prices using ‘appropriate’ CPIs, which • is based on actual price paid (incld. VAT) for exactly the same (group of) product(s); • takes proper account of changes in the quality of products; either adjusted or supplemented by other price data to attain consistency with NA concepts. • ‘B’ Methods: Use of CPIs that are not ‘appropriate’ or adjusted PPIs. • Recommended Methods: Deflation of goods consumption by appropriate CPIs and of services consumption by the deflator used for Production-side estimates at constant price.

12. GFCE and Final Consumption of NPISH (1) • For sequence of accounts, it is necessary to distinguish between ‘individual’ and ‘collective’ components of GFCE. (Final consumption of NPISH entirely is considered to be ‘individual’ consumption.) • The consumption expenditure of the General Government and NPISH = output of non-market g&s – receipts from the sales of non-market g&s + g&s purchased and transferred to households, un-transformed. Individual Component of GFCE • For the first two components, constant-price estimates can be derived extrapolation of the base year value by output volume indicators used for production-side estimates. [see Handout]

13. GFCE and Final Consumption of NPISH (2) • For the third component, ideally the commodity-specific CPIs should be used as deflators. • But, estimates of the third component (usually has an insignificant share) are often not available by products. Thus, the CPI for ‘food & clothing’ may be used as deflator. • Other alternative: Deflation of output at current prices by a compound price index for total observed input of production factors. Collective Component of GFCE • Only option is the “Other alternative” stated above.

14. GFCF at Constant Prices • Gross fixed capital formation (GFCF) - tangible and intangible - represents a large range of different products. • Thus requires deflation at a detailed disaggregated product level for estimating GFCF at constant prices. • At least, separate estimates are required for: • buildings and other construction works, • machinery and equipment, and • motor vehicles and other transport equipment. • GFCF can be measured from either the supply or demand side. • The supply side approach – being less resource intensive - is more commonly used as the main method.

15. GFCF in Buildings and other construction work • The constant-price estimates of GFCF is derived as part of the derivation of estimates for output for the construction industry. • Available methods: • Extrapolation by volume indicators like square meters of finished construction work. • Deflation by output price indices for buildings & constructions – ‘hedonic’ approach or ‘model pricing’ or ‘specification pricing’. • Deflation by input price indices for buildings & constructions. Deflation by implicit price deflator of Construction industry. • Extrapolation by volume indicator based on factor inputs - CE, IC and CFC at constant prices.

16. GFCF in Machinery and Equipment GFCF in machinery and equipment and motor vehicles and other transport equipment - Preferred method: • Compilation at the most disaggregated level available. • Appropriately weighted price deflators like (i) Composite (Paasche) PI of PPIs of relevant components,. (ii) UVI for imports of capital goods. • Appropriate weights: Shares of domestic supply to the domestic market and imports shares of total supply to the domestic markets of each type of capital good.

17. Change in Inventory (CI) at Constant Prices (1) • As per 1993 SNA, for each individual groups of goods, Changes in inventories at current prices = changes in inventories at constant price + nominal holding gains on inventories. • These should ideally be derived from opening and closing balance sheet values. • Alternatively, when supply and use tables are used for compiling GDP - production and expenditure side at current and constant prices simultaneously - changes in inventories at current and at constant prices are commonly derived residually.

18. Change in Inventory (CI) at Constant Prices (2) • Estimating changes in inventories at constant prices by deflating total changes in inventories at current prices is not possible. • Further, period to period rates of changes in (change in inventories) is not very meaningful. • No meaningful volume index for changes in inventories can be defined or constructed.

19. CI at Constant Prices– A/B/C Methods • ‘A’ Method: Requires (i) direct quantity information or value information (ii) knowledge about the bookkeeping system and (iii) appropriate price information, all at quarterly or monthly periodicity, when prices & quantities vary considerably within the year – for calculating CI and holding gains. • ‘B’ Methods: Methods of calculating CI and holding gains when the above requirements are not met. • ‘C’ Methods: Deflation by proxy or inappropriate indicators or use of commodity flow method without any direct data. • ‘Worst’ Method: CI in constant prices is calculated as a residual at the macro-level.

20. Imports & Exports at Constant Prices (1) • Usually, the estimates of exports and imports at constant prices are obtained by deflating current price estimates by UVI. • The data required for UVI - the value and quantity of shipments – are available from declarations the importers and exporters are required to make. • But, unit-value indices are incapable of accounting for quality changes.

21. Imports & Exports at Constant Prices (2) • Thus, export and import price indices (XPI & MPI) are much preferred in comparison to UVI. • But, compilation of XPI and MPI require detailed specifications of products exported and imported, which are normally not available from the Customs department. • A second-best and operationally feasible alternative: Using UVIs based on the finest level of commodity detail available, disaggregated by country of origin/destination.

22. Imports & Exports - A/B/C Methods • ‘A’ method: Deflation by quality adjusted XPI & MPI covering all exports and imports, which are are required to be • consistent with the product classification • based on respectively f.o.b & c.i.f prices. • ‘B’ methods: (i) Deflation by XPI & MPI without quality adjustment. (ii) UVIs based on product groups that are sufficiently homogeneous over time. • ‘C’ methods: All other methods, including using UVIs for insufficiently homogeneous product groups.

23. Components of GDP at Constant Prices Components of GDP: • Compensation of employees (D.1) • Mixed income (B.3) + Operating surplus (B.2) • (Taxes – subsidies) on production & imports (D.2–D.3) • Consumption of Fixed Capital or CFC (K.1) Of these, Mixed income + Operating surplus can be obtained as residual, viz. GDP – CE – (t – s) – CFC.

24. Compensation of Employees(CE)at Constant Prices • CE includes • wages and salaries – in cash & kind • social contributions, and employers' social contributions. • However, deflation of individual elements of CE is not necessary. • It is sufficient to apply the method appropriate for wages and salaries in cash to the entire CE.

25. CEat Constant Prices -A/B/C Methods • ‘A’ method: Extrapolation by an "hours worked" measure separately for sufficiently detailed skill levels of employees. • ‘B’ methods: (i) Extrapolation by “hours paid” without sufficiently detailed occupational classification, (ii) Deflation by average wage or wage rate if estimated at a sufficient level of stratification to distinguish reasonably homogenous groups of employees. • ‘C’ methods: (i) Extrapolation by number of employees, (ii) Deflation by an unrepresentative or general wage rate index or CPI.

26. Taxes & Subsidies at Constant Prices • Distinction required between • (Taxes-subsidies) on products & imports • “Other” (Taxes – subsidies) on production • The later is considered as a component of value added (at basic price) • However, both are required for obtaining the balancing item “Mixed income + Operating surplus”.

27. Taxes & subsidies on products & imports • In general, taxes & subsidies on products at constant prices can not be estimated by deflation. • Taxes & subsidies on products at constant prices for both ad valorem taxes and for quantity taxes, can be obtained by applying the implicit base year tax rates on the relevant transactions at constant prices.

28. “Other” taxes and subsidies on production • Different categories of “Other” taxes and subsidies on production • Payroll taxes – percentage of salary bill; • Taxes on ownership or use of an asset – number, size or value of assets; • Taxes in the form of licenses – at a flat rate • Subsidies on certain processes (eg. reducing pollution) – at a flat rate for qualifying firms. • Administrative records are the main source of data on “Other” taxes and subsidies on production classified by types.

29. “Other” taxes and subsidies on production – A/B/C Methods • ‘A’ method: Applying the base year rates appropriately on the relevant number or deflated value separately for each category. • ‘B’ methods: Same method as ‘A’, but with less detailed data or inappropriate indicators for deflation. • ‘C’ methods: None specified in the Handbook. • A ‘last resort’ method: Deflation by GDP deflator.

30. CFC at Constant Prices • CFC at current prices are estimated from stock of fixed assets and their probable economic life (PIM) – closely linked to estimation of GFCF. • For estimation of CFC at constant prices, the method applied for constant-price estimation of GFCF should be applied for respective types of assets.

31. ﺸﻛﺮﺍ