Philippines: Treatment of Remittances in NTA Rachel H. Racelis J.M. Ian S. Salas

1. Philippines: Treatment of Remittances in NTARachel H. RacelisJ.M. Ian S. Salas SKKU, Seoul, Korea 5 Nov 2007

2. Introduction • NTA aggregate control for labor income: YL = (2/3)*household operating surplus + compensation of employees But, Compensation = compensation from residents + compensation from ROW - compensation to ROW • Age profile of compensation from/to ROW may be different from those of residents.

3. YL components, 1980-2004 Billion pesos

4. Share of YL components, 1980-2004 % share of YL

5. Net compensation from/to ROW • We try to construct age profiles for net compensation from ROW, but remember that this has inflow and outflow components. • Since 1997, however, Philippine NIA indicates zero compensation to the ROW. • If compensation to ROW is significant, how do we construct age profiles for it?

6. Methodology assumptions • While NIA has compensation of non-resident employees (overseas worker), what we can get from survey at this time is compensation net of consumption, assuming that the overseas worker: • did not receive transfers (public and private), • did not acquire nor sold assets, • did not borrow or dis-save, and • sent remittances to only one household at home. • The constructed age profile would be valid if the proportion of consumption to income is the same for all overseas workers.

7. Available surveys • Labor Force Survey (LFS), quarterly • Jan, Apr, Jul, Oct; past-week activity • Individual-level • Annual Poverty Indicators Survey (APIS), every Oct of non-FIES year (triennial) • rider to Oct round of LFS, 6-month coverage • Household- and individual-level • Survery of Overseas Filipinos (SOF), annual • rider to Oct round of LFS, 6-month coverage • Individual-level; respondent is reporting household member

8. Survey details • APIS excludes overseas Filipinos in its household roster. • SOF coverage: • overseas Filipinos who went abroad within the past five years (if >5 years, considered “immigrant”) • asks if overseas Filipino worked abroad • asks the reason for leaving of overseas Filipino and date of return or expected return • n:1 but not 1:n relationship between overseas Filipinos attached to households

9. Some notes on 1999 estimation • 97% of contract workers left five times or less during the previous five years. • Indicates that most contract workers may have yearly renewable contracts. • APIS: 37,454 households; 188,671 individuals • SOF: 1,897 households; 2,174 individuals • In the following, suffixes _h and _m refer to household-level or individual-level data, respectively.

10. Data items used: APIS_h • giftin_abroad_h: “cash receipts, gifts, support, relief and other forms from abroad for past six months, in cash” • may include remittances received from overseas Filipinos which are not attached to the household (immigrant relatives, friends, etc.) • No usable item in APIS_m

11. Data items used: SOF_m • remittance_m: “cash remittance received” + “cash brought home”

12. Data items used: SOF_m • “reason for leaving the country” • Encoded answers: • Contract worker • Work with Phil. Consulate/Embassy abroad • Worker other than contractual • Tourist • Student • Immigrant • Official missions • Others • Different treatment relative to survey definition of overseas worker (arrows + employed).

13. Data items used: SOF_m • “had work/job last stay abroad”

14. Reintegration of overseas Filipinos • Merge household-file and member-file of APIS, then merge this with SOF_m. • Use individual weights from both surveys (each calibrated to correspond to census numbers). • Weights of members of household may be different from overseas Filipino/s attached to it (true for 1999 but not for 2002).

15. Overseas worker definition • Overseas Filipino must satisfy both of the following for remittance to be considered as net compensation from ROW: • be away from his/her household temporarily (to establish continuing ties as member of the household) • be employed (for remittance to be considered as labor income) • Otherwise, remittance will be classified as inter-household transfer from ROW. • remittance_h = comp_ROW_h + inter_ROW_h

16. Adjustments • Ideally, remittance_h (SOF) == giftin_abroad_h (APIS) • If >, scale remittance_h to giftin_abroad_h and apply to comp_ROW_m and inter_ROW_m accordingly. • If <, residual is treated as additional inter_ROW_h. • Is residual attributable to households receiving remittances from “immigrant” overseas Filipinos (not covered by the SOF survey), or is it due to reporting error?

17. Remaining issues • Aggregate control to use for inter-household transfers from ROW (in practice, does net compensation from ROW in NIA exclude remittances unrelated to labor income?) • Survey representativeness: Only 1,897 households have overseas Filipinos, as defined in the survey (some of which were not employed abroad), while 4,682 households receive cash gifts from abroad, so that inter-HH transfers much bigger than labor income of overseas workers.

18. Some notes • Overseas household member sometimes designated as household head, so that resident household head may only be a “substitute” head • we still used “substitute” as head in those cases • Intra-household inflows of overseas workers are zero but outflows are positive as they directly contribute to disposable income (since labor income is already net of consumption).

19. Preliminary age profiles of labor income, 1999 Jan 2007

20. Age distribution of overseas and locally employed workers, 1999

21. Normalized age profile of per worker compensation, 1999

22. Per capita consumption and labor income with and without net compensation from ROW, 1999