The Social Accounting Matrix (SAM)

Finally, we filled the SAM form partly by the above mentioned data, partly by the income distribution data of the national accounts (NA), balance of payments statistics (BP) and the Government Budget Reports (BR). Here some minor methodological problems about the necessary content of certain cells and the placement of several unusual categories still remain.

For example, the export subsidies could not be put into the column of the branches since the original (I-O table) figures of the exports already contain them. The related statistical problem is that the Budget Reports (BR) and the NA show different figures for the aggregate export subsidy (perhaps it is only due to the fact that the BR usually publish cash-flows, while the NA follows the „due payment” or accrual approach). Neither source publish any break-down of this subsidy (from the 1998 BR we know that in that year 77 % went to the food industry and 23 % to the agriculture). As a solution, we accounted the export subsidies as a direct transfer from the government to the foreign sector (as a consequence its direct relationship with the export is not visible). Also there are further minor inconsistencies between the data. For example, it is worth mentioning the inconsistency between the ’dividend to the foreign sector’

figures in the National Accounts and the Balance of Payments statistics. Also for the government net interest expenditure and (theoretically meaningful) deficit we could derive somewhat different figures from the BR and the NA.

The greatest problem was the unclear content of various budget transfer payments and their relationship with the NA categories. It was very difficult to make sure that double accounting is not made. Fortunately, we did not have to deal seriously with the capital and investment transfers, since the model will not deal with these explicitly. Interest payments are not explicitly taken into account either. Therefore, the official figures for savings had to be modified only by them to get the current (or primary or operational) savings.

Table 6.1: The content of the income distribution of the 1995 Hungarian SAM (SAM18HU3.XLS file’s SAM18 worksheet)

Cell name Category

Content (in Million HUF) and source (NA=National Accounts, BR=Budget Report)

AA31 Households' other transfer payments 361326 (NA) other transfer outlays (routed through firms) AA33 Households tax and duty payments 411965 (NA) (of which 383744 was income tax)

AA34 Households social security contribution

894523 (NA)=698879+31007 employer (incl. imputed) +140478 employee+24159 other

AA47 Saving of households net of interest income NA, 598640 saving-138790 net interest income

AB30

Households' cash and in-kind benefits (routed

through gov.!) NA: 910459 cash benefits+763186 in-kind benefits

AB31

Non-export subsidies not accounted as production subsidies

=BR (1500+7302 rehabil.+433 defunct+5580 guarrantee +20384 extrabudg.funds) -29629 (NA) +42428 balancing AB45 Net foreign transfers of the government BoP: 13*125,69

AB47

Primary saving of government (incl.

inv.transf. by LG,EBF too)

=-361100 Flow of funds data by NBH+AJ23 (investment) +(200100+138790-170450+AC23) net interest (residually) +(79247-37546)+12066+(28679-14141) inv. transfers -2500 coal correction-17649 balancing

AD30

Households transfer and property income

without interest NA:485354 transfers+(196726-138790) property income

AD33 Firms' income tax, dividend to state

NA: 92301+22206 income tax of enterprises & banks +10912 dividend

AD45

Dividends,other transfers +discrepancy between the BoP and SAM for. trade acc.

24200+(6700-3500+1500)-(143300+218500-130000 -301400-23300+25400)+(AG24+AG36-AB46-V46)

AD47

Saving of firms & banks, net of interest (incl.reinvested profit)

370196+97547+(99572+26900) reinvested-(AC23-170450) +2500 correction of coal sector’s saving

AE47

Saving of the foreign sector net of interest

income =296500 current account deficit -200100 net interest

AF36

Zero (by GEM-E3 model’s assumption.

Alternatively: Tourist export's VAT content) may be estimated only

Y30 Households non-labor income of production 279533+720149 (NA), operating surplus+mixed income

Y31

Firms' capital income (enterprises, NPIS, banks respectively)

908765+2015+153609 (NA) op.surplus-16921 net other taxes (rent on mining) (BR)+2500 corr. in coal sector Y39 Government capital income 248182 (NA), operating surplus of government institutions

The following comments have to be made for the SAM cells presented in Table 6.1:

AA31: An unpublished background worksheet of the HCSO showed that of the NA accounted 361326 million HUF transfers paid by the household 83 per cent (299942 Million HUF) were ‘transfer expenditures in foreign currency’. The rest were insurance (43685) and gambling (8186). Similarly, in

AD30 of the 485354 received transfers of the households 90 per cent (436627) was

‘transfer income in foreign currency’. The rest were insurance (40541) and gambling (8186, apparently accounted as income received from other households).

AA34: the ‘other’ SSC consists of SSC of the self-employed, the voluntary SSC and SSC after people on unemployment benefit, child care leave and sick leave.

AB47: Non-households investment transfers were not published in the NA before 2002.

Hence the data are estimated from the Budget Reports.

AD30: Household other property income consisted of dividend (35432), insurance (15132) and (land) rent income (7372).

AD45: Data for the net dividend income of the RoW are rather different in the NA (58535-3665=54870) and in the BoP (24200=‘direct investment income’, where reinvested profits were not included here yet at that time).

AD47: Since in the NA savings are accounted after subtracting reinvested profits, here we had to add them back (since in the transfer cells of the SAM we did not account them for). For the coal correction see Y31.

AE47: In contrary, in the BoP the current account deficit (i.e. the mirror image of the savings of the RoW) was accounted without computing the reinvested profits. Therefore we did not have to add it back to the BoP figure for the savings of the RoW.

AF36: If the SAM displays the domestic consumption, then no VAT should appear here.

Even if the SAM displays the residents consumption, and hence the exports include the consumption of inbound tourists, the GEM-E3 model assumes does not distinguishes between the 2 types of the exports (enterprise and tourist), so it assumes that all VAT is accounted for in the column of household consumption. If the SAM’s export data are at user’s prices (as in the NA), VAT on foreign tourists’ consumption remain unseparated by definition. However, if I-O table data are used, which usually at basic prices, the difference between users process and basic prices (incl. export taxes and subsidies) have to be accounted for somewhere in the accounts of the RoW. In any case, in our SAM these differences (6504 tax-54712 subsidy=-48208 net tax, i.e. negative net tax) are added to the basic prices. It however means that exports are not really at uniform (users) prices, since foreign consumers will have to pay the foreign VAT and other consumer taxes when purchasing the products exported by Hungary.

Y31 and AB47 and AD47: The capital income of the coal sector seemed negative in the Hungarian statistics. However, it is impossible that the real cost of capital was negative.

Therefore we had to find in the Budget Report that yet unaccounted government transfer which helped the survival of the coal sector. However, note that these bailing out transfers usually

take rather hidden forms, ranging from write-offs and preferential interest rates to equity lift-offs.

After filling the above block of the SAM without the residual corrections (indicated by the bold letters in Table 6.1), total savings and total accumulation did not match exactly, but the difference was surprisingly small (at least relative to what one would expect after learning the above mentioned methodological inconsistencies and coverage problems). We concluded that the remaining error certainly must be in the accounts of the government. It is not only because of the above mentioned uncertainty in the interest expenditures, capital transfers and savings, but also because of the incomplete or not guarranteedly correct accounting of certain transfers between the firms and the government. For example, the (although small amount of) transfers of the „non-profit institutions” are not necessarily properly accounted for, since their account in the SNA is also incomplete and the government budget data do not separate them out.

However, by determining some cells as residual (see the residual corrections in Table 6.1 in bold letters), the SAM could be balanced easily. Table 6.2 shows the aggregate balanced SAM.

Table 6.3 shows the aggregate balanced SAM converted to Euros and after eliminating the FISIM (as mentioned in section 5.3. and 5.4.1.) We think the resulting SAM can be used by the model well, but if any further corrections should prove to be necessary, the elaborated framework will make it easy to do.