EUROSTAT
Directorate E: Social and regional statistics and geographical information system
Unit E-2: Living conditions /
Doc. HBS/2000/129/00
Expert Group Meeting
Household Budget Surveys
13 December 2000
Bech Building - Room B2/464
Agenda item IX
1
TREATMENT OF EDUCATION AND HEALTH EXPENDITURE IN THE ECPF (SPANISH CONTINUOUS HOUSEHOLD – SURVEY)
INTRODUCTION
When the change of the continuous household budget survey was thought of, not only the recommendations made by EUROSTAT were taken into account in connection with the new needs of the national accounting system and the social indicators systems but also the lack of information that existed in some fields of social interest.
This last objective shall be attended through the introduction of specific modules in the survey in those periods in which there is a weaker collaboration of the families and when it is made possible given the techniques and financial support of the project.
So, it can be said that the survey is a correct tool to investigate the different variables that are not available in the now existent sources of information, but only in a partial way, limited by their main objectives. Specialised investigation about some fields, such as health or education will still need ad hoc statistical operations.
Trying to integrate the surveys into the statistical system involves the need to use the same definitions, concepts, classifications, etc.… as in the other operations of system.
This impelled the main objective in the new HBS to be the measurement of real final consumption of households as proposed by national accounts, instead of how it was done up to now, i.e. with the concept of final consumption expenditure of households.
So the purpose is to add to the final consumption expenditure the value of expenditures (social transfer in kind) made by the public administration or by non-profit institutions in goods and services given to the households for their consumption without any transformation, mainly in the fields of health and education.
A total approximation is not possible, due to the difficulty that the families would have to estimate the value of those goods and services that are given for free or at a lower price than in the market. None the less, in health as well as education what is wanted is to make a small approximation to consumption through the use of the services.
Therefore and for the last 12 weeks previous to the sample week (the survey is taken quarterly), the number of times during that period each and all of the members of the household visited the doctor (general or specialist), and dentist (logically the visits to ask for prescription slips are not included), the amount paid for the visit if the family made any payment, and the percentage reimbursed on the amount paid are collected. This same information is pick-up in case of visit to the annalist and radiologist but the reference period is monthly. For hospitalisation similar information is captured, for the last 12 weeks, adding to them the number of nights of stay.
The drugs, pharmaceutical services and acquisition of proteases, therapeutically equipment, etc. Have a similar treatment, with a reference period of 1 month.
This information cross-referenced with the type of health coverage[1] (which originally was taken for each one of the members of the household, now is only collected for the reference person), show us who has provided the health service, and would make possible to assign an expenditure to the families user of such services, either through external sources, or without considering the coverage variable, based on the expenditure derived from the survey itself.
In this way, we shall have information not only of final consumption expenditure on health, but it shall also be possible to identify the group of users for the social protection services, and therefore know the influence that it could have on society the reduction of the said services, or what needs would be rendered necessary.
The knowledge of the use of the educational services is done through the type of centre where the individuals carried-out their studies[2]. This information joined to educational level would allow assigning expenditure to the families that use the free services, through the data extracted from the survey itself, and learning the characteristics of the user of such services.
USE OF PUBLIC HEALTH AND EDUCATION SERVICES IN SPAIN.
In a country like Spain where there is an universal health coverage, and therefore the family expenditure concerning health should be limited exclusively to the contributions made to the Social Security and trough the payment of taxes, we find that a part of the households being protected by the Social Security system, choose an alternative system for health coverage and even go to the private health sector when they consider it necessary. So if we want to compare these households with those which exclusively use public health, for example, we would have to assign to those last ones an expenditure.
TABLE 1. TYPE OF HEALTH COVERAGE[3]
TYPE OF COVERAGE / NUMBER OF HOUSEHOLDS / PERCENTAGE OVER TOTALONLY SOCIAL SECURITY / 10.006.276 / 83,2
ONLY PUBLIC INSURANCE (HEALTH) / 576.163 / 4,8
SOCIAL SECURITY AND PRIVATE HEALTH INFORMANCE / 1.262.666 / 10,4
PUBLIC HEALTH INSURANCE AND OTHERS / 57.125 / 0,5
OTHERS / 125.398 / 1,1
TOTAL / 12.027.718 / 100,0
Source: ECPF 2nd quarter 1998
In table 1, it can be observed that 11% of the reference persons have, besides public health coverage, an alternate source of social protection.
TABLE 2. TOTAL EXPENDITURE, REIMBURSED AMOUNT, AND NUMBER OF VISITS PER TYPE, ACCORDING TO MEDICAL SPECIALITY
NUMBER OF VISITS / TOTAL (Thousands) / WITHOUT REIMBURSEMENT / WITH REIMBURSEMENT / ALL FREE / AMOUNT (Millions) / REIMBURSED AMOUNTMEDICAL SPECIALITY
GENERAL PRACTITIONER / 70.229 / 70.084 / 144 / 68.890 / 4.562 / 73
ESPECIALIST / 52.570 / 52.365 / 204 / 42.336 / 79.426 / 415
DENTIST / 15.045 / 14.954 / 91 / 2.176 / 183.005 / 1.102
ANALISIS OR RADIOLOGIST / 20.762 / 20.721 / 42 / 18.209 / 20.071 / 213
NOT AVAILABLE / 50 / 50 / - / 50 / - / -
TOTAL / 158.656 / 158.174 / 481 / 131.661 / 286.884 / 1.803
Source: ECPF. Year 1998 provisional results.
It can be noticed in table 2 the use of private health services. Of the 158.656.113 medical visits made in the year 1998, 16,4% (some 26.000.000) were private medical visits and therefore totally paid by the households. The highest percentage appears in the case of the dentist since it is precisely where the social protection system is weaker.
Hospitalisations the outcome change substantially; only 4,2% of the hospitalisations are paid totally by the families, the rest of services being given by public systems of private insurance companies.
Otherwise, and as it was mentioned before from the survey groups of users of social services can be identified (even though a perfect identification cannot be made, it would be necessary to separate in case an individual could have access to coverage either by the SS or by a private insurance, which one of them he was using).
Table 3 shows information on the percentage of individuals inside each age group that most frequently use the free services over the total number of visits.
TABLE 3. PERCENTAGE OF FREE VISITS OVER THE TOTAL NUMBER OF VISITS BY MEDICAL SPECIALTY ACCORDING TO AGE GROUP
TOTAL / GENERALIST / SPECIALIST / DENTIST / ANNALIST OR RADIOLOGISTUp to 20 years / 77,84 / 96,56 / 81,39 / 16,43 / 88,16
From 21 to 40 years / 73,56 / 97,84 / 72,48 / 12,82 / 84,91
From 41 to 64 years / 83,70 / 98,63 / 84,45 / 13,34 / 89,96
More than 64 years / 92,59 / 98,89 / 88,41 / 20,81 / 88,93
Source ECPF 2nd quarter 1998
As it could have been expected the elderly people are the main users of such services.
Concerning the level of income in the household (see table 4), the households with less income are the ones that use more the free services; being less used by those households with a higher standard of living.
TABLE 4. PERCENTAGE OF FREE VISITS OVER THE TOTAL NUMBER OF VISITS BY MEDICAL SPECIALITY ACCORDING TO THE HOUSEHOLD INCOME
HOUSEHOLD INCOMES / TOTAL / GENERAL PRACTITIONER / SPECIALIST / DENTIST / ANNALIST OR RADIOLOGISTUp to 65.000 / 94.43 / 98.41 / 92.08 / 21.75 / 98.66
From 65.001 to 130.000 / 88.87 / 99.26 / 85.78 / 18.96 / 81.86
From 130.001 to 195.000 / 85.18 / 98.99 / 85.02 / 14.28 / 93.55
From 195.001 to 260.000 / 80.43 / 97.71 / 79.51 / 12.18 / 93.02
From 260.001 to 325.000 / 78.10 / 96.98 / 81.43 / 15.79 / 89.83
From 325.001 to 390.000 / 73.54 / 95.52 / 77.97 / 15.75 / 77.73
From 390.001 to 650.000 / 64.14 / 95.73 / 65.53 / 8.65 / 75.92
More than 650.000 / 49.31 / 88.02 / 52.74 / 2.41 / 100.00
Source: ECPF 2nd quarter 1998
The measure of the real final consumption has its advantages: it makes international comparisons easier, and between groups of households it filters the user’s of the services… nonetheless and as it can be seen, in literature on the subject, it improves artificially the living conditions of the populations with worse health status and their relative position respect to the more healthy groups. The data extracted from the survey support this hypothesis.
Table 5 shows the change produced in the household when they are classified by the deciles of final consumption and by the deciles of consumption expenditure.
TABLE 5. DISTRIBUTION OF THE HOUSEHOLD ACCORDING THE REAL FINAL CONSUMPTION DECILE AND ACCORDING THE DECILE OF FINAL CONSUMPTION EXPENDITURE
Real Final Consumption DecileDecile / Decile 1 / Decile 2 / Decile 3 / Decile 4 / Decile 5 / Decile 6 / Decile 7 / Decile 8 / Decile 9 / Decile 10
Decile 1 / 89,14 / 8,69 / 0,92 / 1,02 / 0,23
Decile 2 / 10,81 / 78,48 / 8,12 / 1,55 / 0,68 / 0,35
Decile 3 / 12,82 / 75,27 / 8,82 / 2,40 / 0,53 / 0,15
Decile 4 / 15,64 / 72,63 / 9,66 / 0,96 / 0,69 / 0,43
Decile 5 / 15,84 / 71,58 / 9,32 / 1,97 / 1,29
Decile 6 / 15,57 / 74,96 / 8,04 / 1,43
Decile 7 / 13,74 / 76,66 / 8,19 / 1,41
Decile 8 / 12,35 / 79,98 / 6,94 / 0,74
Decile 9 / 9,09 / 85,84 / 5,07
Decile 10 / 5,78 / 94,22
Source: ECPF 1999 3rd quarter of 1998
Nonetheless, this is not the only problem that the real final consumption approximation can give us. The difficulty for the families to take note of each and all of the visits to the doctor[4] whether they are free or not, in an already complicated survey is other important problem. The approach to the real consumption even though very interesting[5] to turn out to be very difficult.
The measuring of the real final consumption approach in education is easier than in the health case, because the measuring of this type of consumption does not produce the distortions in the standards of living seen in the health consumption case, nor it is subject to the difficulties implicit in the measuring of the use of the health services.
According to the data obtained from the HBS 79% of the people that study do it in public education institutions, 13,8% do it in private subsidised education institutions and only 7,2% study in private non subsidised institutions. In order to compare the data from different households it would be necessary to impute an expenditure to the users of the public or private subsidised education, or otherwise using information from the educational expenditure data given by the survey[6], or also from external sources ( expenditure financing survey, education accounts, Education Ministry, etc.).
In a similar way in the case of the use of the public health services, in the educational case we can find differences between the population that use these services and the ones that choose other alternatives.
As it is shown in Table 6 the percentage of expenditures on food is higher in families that receive education in public institutions, and resulting that the percentage of expenditures in restoration is higher in families that are users of private education.
TABLE 6. PERCENTAGE OF EXPENDITURE OVER TOTAL EXPENDITURES, ACCORDING TO GROUPS OF EXPENDITURES
Type of centre / Group 1 / Group 11Public / 18,27 / 10,00
Private / 13,78 / 11,24
Source: ECPF 2nd quarter 1998
AN ALTERNATIVE TO THE MEASURMENT OF HEALTH AND EDUCATION CONSUMPTION
The Richness of information provided by the measurement of “real final consumption”, is overshadowed in the case of health due to the difficulty that a continuos register of each one of the visits to the doctor and the hospitals stays suppose for the families, above all in the Spanish HBS where the families participate in the survey during eight consecutive quarters. As it has been already said, an under estimation of the use of health services has been noticed when the data from the HBS are compared with external sources.
This problem is not observed in the case of use of educational service, so when a decision has to be taken on how to estimate the expenditure-consumption in the HBS a different measurement is propose in the health field and in the education one.
The proposed approach for the estimation of the health expenditure is that of the “cost of disposability”, that is, the pay out that the families have to make in order to be able to use the health services. In order to obtain the estimation for the health total expenditure, we have to add to the cost of disposability the health expenditures for the families.
The “cost of disposability” is given by:
C · D = CSS + PBSS + (%I)[7]
Where
CSS are the contributions to the Social Security or similar payments.
PBSS premium for private insurance.
I% percentage of taxes assigned to health expenditure.
The health final consumption expenditure (G. R.)
Would be obtained
GR = D – R
Where
D = Non refunded expenditure in health goods and services.
R = Expenditure refunded at a later date by the social security organisms, by non profit institutions or by private companies.
So, to calculate the real final consumption (G. T.) in health it would be enough to just add the components above:
GT = CD + GR = CSS + PBSS + (%I) + D - R
This way of looking at it allows the international comparison, it measures in conjunction with real expenditure the cost of disposability of the health services and has the advantage in comparison with the real final expenditure that it is of easier implementation and it does not artificially distort the position of the living standard in the sick population.
It has disadvantages, that the health expenditure can not be divided according to the type of expenditure (drugs, medical visits, hospitalisations…) and that it assigns a disposability cost of zero to the persons that are exempt from payment of social payment (retired, unemployed, etc.) precisely the population that make more use of these types of services.
This last obstacle could be worked around by breaking-up the total expenditure in contributions during the whole working life.
The proposed method for the measurement of the real final expenditure in education would be the one obtained from the use of educational services since as it has been said beforeis easy to apply and it is considered necessary when doing international comparisons.
ANNEX 1
TABLE 3B. NUMBER OF FREE VISITS TO THE DOCTOR BY MEDICAL SPECIALITY ACCORDING TO THE AGE GROUP
GENERALIST / SPECIALIST / DENTIST / ANNALIST OR RADIOLOGISTUp to 20 years / 2.501.524 / 2.716.382 / 175.706 / 438.006
From 21 to 40 years / 2.210.334 / 1.827.447 / 130.081 / 694.709
From 41 to 64 years / 5.474.594 / 3.489.605 / 181.733 / 1.412.111
More than 64 years / 6.934.635 / 2.577.722 / 81.225 / 987.587
Source ECPF 2nd quarter 1998
TABLE 4B. NUMBER OF FREE VISITS TO THE DOCTOR BY MEDICAL SPECIALITY ACCORDING TO THE HOUSEHOLD INCOME
HOUSEHOLD INCOME / GENERAL PRACTITIONER / SPECIALIST / DENTIST / ANNALIST OR RADIOLOGISTUp to 65.000 / 1.661.810 / 632.413 / 19.171 / 302.957
From 65.001 to 130.000 / 5.537.856 / 2.601.863 / 118.019 / 621.172
From 130.001 to 195.000 / 4.314.979 / 2.843.231 / 138.336 / 1.005.055
From 195.001 to 260.000 / 2.536.981 / 1.815.367 / 89.893 / 561.973
From 260.001 to 325.000 / 1.073.345 / 1.161.771 / 77.147 / 359.619
From 325.001 to 390.000 / 582.509 / 499.967 / 46.846 / 177.569
From 390.001 to 650.000 / 584.861 / 528.612 / 36.883 / 209.322
More than 650.000 / 58.294 / 50.942 / 13.417 / 21.648
Source: ECPF 3rd quarter of 1997.
1
[1] Type of health coverage
- Public Social Security
- Public insurance health
- Private health insurance
- Others
[2] Type of centre of studies
- Public Centre in National Territory.
- Private Centre, subsidised, in National Territory.
- Private Centre, not subsidised, in National Territory.
- Centre in Foreign country.
[3]The data refer to the Reference Person
[4] Testing the data obtained from the HBS against external sources, an underestimation of the number of stays (number of nights) in the hospitals has bees observed.
[5]The capacity to break-up the real final consumption, into the final consumption expenditure and the transfers in kind would make possible when thought necessary to add the transfers with the expenditure, and even to allow o sensibility analysis with alternative measures.
[6]Due to small percentage of students in private education, the estimation through data from the survey would make necessary to accumulate the sample from different periods.
As for as possible is always prefer to imputed the use of free educational services through data given by the survey itself.
[7]Due to the difficulty to measure the percentage of taxes dedicated to the health care this component could be eliminated.