Malta - Change in Inventories

Malta: Change in Inventories

Mnemonic CIVT.IMLT
Unit Ths. EUR, NSA
Adjustments Not Seasonally Adjusted
Quarterly 47.24 %
Data 2018 Q4 -39,350
2018 Q3 -74,577

Series Information

Source National Statistics Office (NSO) - Malta
Release National Accounts
Frequency Quarterly
Start Date 3/31/2000
End Date 12/31/2018

Malta: Business

Reference Last Previous Units Frequency
Capacity Utilization 2019 Q2 78.2 77.1 %, SA Quarterly
Business Confidence Apr 2019 -22.4 -2.5 SA Monthly
Change in Inventories 2018 Q4 -39,350 -74,577 Ths. EUR, NSA Quarterly
Industrial Production May 2017 97.3 98.1 Index 2010=100, SA Monthly
Real Change in Inventories 2009 -14,743,020 40,522,741 NCU Annual

Release Information

Malta has compiled annual national accounts since 1954 without fully adopting the System of National Accounts 1968 (1968 SNA). From 1999 to 2014, the 1995 ESA 1995 was followed as the general framework for compiling the national accounts statistics, which is consistent with the System of National Accounts 1993 (1993 SNA) this data was replaced with ESA 2010 data  in late 2014. Currently the data is reported using the ESA 2010 framework.


Malta's Gross National Income (GNI) Inventory is a 356-page document containing the sources and methods used for the compilation of the GDP and the GNI based on the European System of Accounts (ESA 1995).  The document is available online:  The GNI Inventory was last updated in 2008 and makes reference to NACE Rev 1.1.  Users should be informed that the GNI Inventory currently available on our website describes the sources and methods of the ESA 1995 series.  Following the implementation of ESA 2010, all EU Member States have been requested to update their GNI Inventory by December 2015.

Recording basis

With some exceptions, transactions and flows are recorded on an accrual basis. For changes in inventories, no adjustments are made for holding gains.

Work-in-progress in the form of capital formation is recorded in the period the production takes place.

With regard to the government sector, transactions for central government are reported on a cash basis. The adjustments made by the compiling unit in the NSO are based on aggregated information provided by all departments through the treasury department. However it should be noted that accounts of the extra-budgetary units and the local councils are already on an accruals basis.

Grossing/netting procedures

Most transactions between establishments (local types of activity units) within the same enterprise are recorded on a net basis since data are reported at the enterprise level. However, for some large companies, which provide detailed information by establishment and activity, transactions are recorded gross as recommended in the international guidelines.

Source data collection programs

Compilation of Maltese national accounts relies on a number of statistical surveys, censuses, and data prepared within the NSO, as well as administrative data supplied by other agencies.

The Business Register (BR) is used as a common sample frame for all enterprise surveys and is essential in estimating GDP by production. Malta’s BR was created in 1997 based on information from the VAT Register and supplemented by secondary sources provided by the Malta Financial Services Authority. It is maintained on an ongoing basis by the NSO Business Statistics Unit. Its updating relies mostly on information from the Structural Business Statistics (SBS), the monthly updates from the VAT Department, and the information on financial and insurance and renting companies provided by the CBM. Currently, the BR includes 53,000 units, consisting of large and small resident units including dormant unit enterprises.

The units registered are the enterprises; work is ongoing to identify the legal units and enterprise groups (including registration of multinational groups) with expected results soon. Data for enterprise groups are being compiled in compliance with the Eurostat’s regulations. Variables included in the recordings provide information on status, type of activities the unit is engaged in, turnover, investment, and employment data. Enterprises are classified according to the 4-digit level of the NACE.

Statistical surveys

  • The Structural Business Survey (SBS) is the major survey, undertaken annually since 2000. Until 2002, a census was conducted covering all units in the BR. The sample design of the SBS is based on a stratified selection of units using the equivalent full-time working persons. For example, the 2003 SBS used a sample of 11,000 units covering all units employing 10 employees or more and a sample of units employing nine or fewer employees. All the units surveyed are also questioned for foreign affiliate purposes (BOP). The information collected refers to the operating accounts and balance sheets of the units. The response rate was about 67–68 percent (number of responding units) but superior in terms of turnover (over 80 percent), which constitutes good source data for national accounts compilation. Information from the Malta Financial Statistical Register is used to impute for nonresponses.
  • The NSO conducts the Household Budgetary Survey (HBS) normally every five years. The last HBS was carried out between February 1, 2008 and February 28, 2009. It covered a national representative sample of 3,732 households throughout Malta, randomly selected.
  • The Labor Force Survey (LFS) is carried out on a quarterly basis with a specific reference week for the quarter using a random sample of 2,500 households spread across the country. The survey methodology follows the recommendations of the International Labor Organization and accords with definitions of employment and unemployment outlined by the Eurostat for EU member countries. The economic activities in which the interviewed persons work are classified according to the NACE classification and the reference period is a week. Because the survey does not provide reliable information at a disaggregated level, the NAU uses it in conjunction with the data from the Employment and Training Corporation (ETC).
  • The Short-Term Business Statistics (STBS) survey represents the main source data for compiling quarterly accounts for the manufacturing sector. The collected information includes the employment, wages, local and export sales, and investment covering all activities, similar to the SBS.
  • The NSO Agriculture Unit undertakes the Agricultural Census every ten years. The last one was in 2001. However, every two years, a Farm Structure Survey is conducted (last one in 2007). This survey uses a sample covering about 14 percent of the respective population derived on the basis of the economic size of farms. The response rate is 88 percent. The Agriculture Unit also conducts an annual survey on livestock.

Source data definitions, scope, classifications, valuation, and time of recording

For the most part, the source data approximate the definitions, scope, classifications, valuation, and time of recording required. Where feasible, the NAU undertakes the necessary adjustments to approximate the NA concepts. Main areas where adjustments are made concern the following:

  • Supplementary source data are sought, and adjustments are made to capture all types of transactions in the economy, such as payments in kind and other nonmonetary transactions, which are important for NA estimation.
  • In estimating the quarterly accounts, the NAU compilers use alternative sources where accounting data are not available on a quarterly basis or their level of detail is insufficient.
  • Classification of the units in the VAT Register is not fully in-line with the NACE used in national accounts. Before using the tax information extracted from the Register, the NAU cross-checks identification/coding against the BR recordings.
  • Adjustments are made to customs data to capture transactions recorded by two different databases (Intrastat and Extrastat). This is a follow-up to Malta’s joining the EU.

The coverage of total economic activities in terms of value added by the data sources is good.

Source data timeliness

Availability of source data varies according to the sector of activity, most problematic being service area and construction. Source data are incorporated into the national accounts as they become available. The inadequate timeliness of survey processing adversely affects the timely production of the annual accounts and the accuracy of the first quarterly estimates. Long delays in certain survey processing are associated with weak response rates, for which the NSO has to reiterate the request. The annual estimates released are the sum of quarters until annual source data become available, in particular the survey results.

Source data statistical techniques

Before incorporating source data into national accounts, the sectoral specialists treat the data, using sound compilation practices. Using additional external information, they impute for missing units (particularly for small enterprises or self-employed persons) to achieve a full coverage of the sector where quarterly survey results are found weak, VAT reports or other administrative data are employed (see also 3.1.2). The sectoral specialists also compare the basic data with external information for their assigned industries.

Other statistical procedures

Specialist studied the nonobserved economy for the first time in 2000 and used the results in the following years as benchmark data. Some estimates for the informal sector in areas such as agriculture, personal services (doctor, dentist, lawyers, etc.) were captured in the national accounts previously. The main types of phenomena covered by the 2000 study related to the (1) underreporting by enterprises and misclassification of activities in survey/censuses, (2) unregistered entrepreneurs (for tax evasion), and (3) estimates for some of the illegal activities (reproduction of media and illegal gambling). The main activities covered by the study included agriculture and fishing, quarrying, manufacturing, construction, trade, hotels and restaurants, transport, other business, and personal services (notably, services by paid domestic staff). However, at present, drugs are not yet captured in compilation of national accounts.

Production approach procedures

The estimates of output and intermediate consumption are compiled at the two- or three-digit level of the NACE, where feasible. Both output and intermediate consumption are compiled for 60 activity groups.

Source data used to derive the output and intermediate consumption are different according to the industry particularities (see 3.1.1). In general, estimates are based on enterprise financial statements derived from either surveys or administrative sources. Adjustments are made to the indicators¾as recorded in the book keeping¾ to follow national accounts concepts, as well as to ensure a comprehensive coverage by activity. For example, where survey data do not cover small enterprises or the response rates are low for some activities, benchmark data from previous periods are used to estimate the value added.

On the specific issues regarding GDP compilation, the following are noted:

Owner-occupied dwellings:

Housing output is valued using the user cost method, owing to specific characteristics of the Maltese Islands’ rental market. Output is obtained by summing up the expenditure components (intermediate consumption, consumption of fixed capital, net taxes on dwellings, and net operating surplus on dwellings and underlying land).

For some of the cost components, the benchmark data used were derived from the HBS, complemented by other sources. Adjustments were made to avoid double counting and to correct the imputed value on empty dwellings.

Inventory valuation adjustment:

Inventories used to estimate output and intermediate consumption are not adjusted for holding gains or losses.

Consumption of fixed capital:

The calculation Consumption of Fixed Capita (CFC) is based on two approaches. The direct method was applied in the case of Dwellings, whilst the Perpetual Inventory Model (PIM) was applied for all the remaining assets. In case of Dwellings, the stock is first valued at replacement cost and CFC is subsequently compiled by assuming an equal depreciation of dwellings over and average period of the service life of 75 years. The PIM has been applied across sectors i.e. for S.11, S.12, S.13, S.14 and S.15. Specific service lives were applied for sixteen different asset types. Due to the fact that no relevant information on service lives is available by industry, industry breakdown was subsequently done in proportion to the old estimate which was more based on company accounts.

Cash versus accrual:

Taxes and subsidies on products, government revenue, and government expenditure are converted from cash to accrual basis by allocating the transactions to the period to which they relate.

Volume measures of GDP: No volume estimates of GDP by activity have been compiled, however, work on this project is currently progressing.

Expenditure approach procedures

All components of GDP by expenditure are derived independently.

Household final consumption expenditure is generally compiled at a four-digit level of the COICOP. The HICP is used to deflate most of the components of household final consumption expenditure. Expenditures by residents abroad are deflated using the inflation rate of the country visited.

Government final consumption expenditure

is compiled at the two-digit level of the COFOG. The volume estimates are derived using an average price index of the two main components of the expenditure¾compensation of employees (deflated by an index covering the 20 salary scales weighted by the number of employees in each scale) and expenditures on goods and services (deflated by the implicit price index of households final consumption) weighted by their share in the nominal value.

Proper techniques are used to address specifics of expenditure on GDP compilation:

  • Government final expenditure excludes incidental sales.
  • Expenditure of residents abroad is included in estimates of imports.
  • Expenditure of nonresidents in the domestic economy is included in exports.
  • Expenditures on valuables are included in gross capital formation for enterprises and households. They are estimated from the supply side as net imports plus domestic production plus a mark-up.

Volume estimates by category of expenditure are derived using annual chain linking with the year 2000 as the reference year.

Currently, only volume estimates for the expenditure categories of the GDP are produced. Thus, the official estimate of real GDP growth is derived from the expenditure side only. However, for the nominal GDP, the production approach is considered the most reliable method because the series are compiled using more comprehensive and reliable data. This aspect further strengthens the need to produce the volume estimates by type of activity, as a second measure for the GDP growth rate.

Specific quarterly compilation techniques

Compilation of quarterly GDP estimates started with 1972, but 2010 ESA figures are available from 2000 onwards. They are compiled using some of the same data sources as the annual estimates plus the infra-annual surveys and administrative data where feasible.

The quarterly estimates are not benchmarked to the annual estimates, limiting the usefulness of the data for temporal comparisons. However, the quarterly estimates are seasonally adjusted using the TRAMO/SEATS ("Demetra"), an ARIMA model-based method, as recommended by Eurostat. The adjusted series includes GDP estimates by expenditure in current and constant prices. The constant price estimates (same base year as for national accounts 2000) cover all the quarters back to 2000. In conformity with international best practice, the source data used to compile the quarterly estimates are not seasonally adjusted.

In the most recent publication of Malta's National Accounts, the National Statistics Office (NSO) provided the 2011 annual revised data, but did not publish the revised quarterly. Presently, the quarterly data we have for 2011 is the most current provided by the source. 2011 is the only year in the dataset affected by this discrepancy. We hope in the following releases to receive the correct data. It is to be noted, however, that this may not be an isolated incident and to use the dataset with caution.