Comparing the traditional to a general level aggregation and growth decomposition of GDP in constant prices: Application to the Philippines

• Main Author: Dumagan, Jesus C.
• Format: text
• Published: Animo Repository 2022
• Subjects: Real prices; GDP level aggregation; GDP growth decomposition; index numbers
• Online Access: https://animorepository.dlsu.edu.ph/res_aki/2; https://animorepository.dlsu.edu.ph/context/res_aki/article/1003/viewcontent/dlsu_aki_working_paper_series_2022_06_083.pdf
• Institution: De La Salle University

The traditional (TRAD) level aggregation and growth decomposition of GDP in constant prices—presently implemented by the Philippine Statistics Authority (PSA)—do not have real prices. However, a generalized exactly additive decomposition (GEAD) framework illustrated in this paper uses real prices—ratios of sectoral GDP deflators to the overall GDP deflator—to convert the GDP of sectors to the same units for GDP level aggregation prior to GDP growth decomposition. In GEAD, a sector’s contribution equals a pure growth effect (PGE) from the sector’s GDP growth—which equals PGE in TRAD—plus a price change effect (PCE) from a change in the sector’s real price, which could be positive, zero, or negative. However, the sum of PCE for all sectors is zero, so GDP growth equals the sum of PGE in TRAD and GEAD. But the zero-sum of PCE does not justify ignoring PCE because this paper shows each sector’s PCE is statistically significantly different from zero over time—negative for Agriculture and Services but positive for Industry—in Philippine GDP from 2001 Q1 to 2019 Q3. Thus, with no PCE, TRAD yields only PGE, which is misleading because it is just one part of a sector’s growth contribution. Therefore, PSA should consider replacing TRAD with GEAD in practice.