MACROECONOMIC DETERMINANTS OF THE DECLINING

AGRICULTURAL EMPLOYMENT SHARE IN THE PHILIPPINES

(1980-2023): AN ARDL APPROACH

ANGEL MAUREEN M. TENORIO

BACHELOR OF SCIENCE IN ECONOMICS
Resource and Environmental Economics

CHRISTOPHER D. BALUBAYAN, EdD

Adviser
BACKGROUND OF THE STUDY
Agriculture has long been the
backbone of human civilization, evolving
from small-scale farming into a vital pillar of
the global economy. It remains crucial
today, especially in countries where it is a
primary source of livelihood (Rodriguez,
2024).

The share of the world’s workers

employed in agriculture fell from 1.1 billion
or 40% of 2.6 billion workers in 2000 to 884
million or 27% of 3.3 billion workers in 2019
(Food and Agriculture Organization, 2021).
BACKGROUND OF THE STUDY
In the Philippines, the agricultural sector
has historically played a vital role in employment
and economic stability. However, recent data
indicate a persistent decline in the share of
agricultural employment, which resulted in a
significant decline in its contribution to the
economy as well. Once a major driver of growth,
its share of GDP has dropped to less than 10% in
recent years.

While this transition reflects economic

progress, it also raises concerns about rural
employment and food security. Savary et al.
(2020) emphasized how the decreasing
workforce in agriculture threatens global food
supply chains and could lead to higher food
prices and increased vulnerability to food
shortages.
RATIONALE OF THE STUDY

Author and Year Study

Lerman (2017) Wage differential is a key driver of declining
agricultural employment, as workers rationally
move toward higher-paying opportunities when
available.
Thomas and Zarema Land ownership and transfer rights as critical
(2021) institutional factors that affect agricultural
employment transitions.
Cerutti and Li (2021) The agricultural exodus represents a transitional
process where workers move from agriculture to
non-agricultural sectors, often experiencing an
Contradicts simplistic models of
agricultural employment
intermediate phase of unemployment or non-
transitions. They noted that participating.
more complex labor dynamics
are still at work.
OBJECTIVES OF THE STUDY

This study aims to analyze the dynamics of the declining agricultural employment
share in the Philippines by using the ARDL Approach. Specifically, this study aims to;

1. To present the trend of key economic indicators in the Philippines, specifically GDP
per Capita, Share of Industrial Gross Value added in GDP, Share in GDP of Net
Foreign Direct Investment Inflows, Share in GDP of credit to the private sector, and
Agricultural Employment Share.

2. To present the short-run and long-run influence of these indicators on the Agricultural
Employment Share of the Philippines over the period 1980 to 2023.
 SIGNIFICANCE OF THE STUDY
This study would be beneficial for the following:

• Department of Agriculture (DA) • Agricultural Workers

- research findings can be used as a guide in formulating - can provide valuable insights into the long-term
strategies and policies to enhance agricultural productivity trends affecting the agricultural sector.
and efficiency despite a shrinking labor force. • Policy Makers
• Department of Labor and Employment (DOLE) -This study contributes to policy development by
- can use the research findings to mitigate potential recommending effective agricultural policies that
unemployment, manage rural-urban migration, and foster could improve agricultural employment in the
workforce adaptability. Philippines.
• National Economic and Development Authority • Future Researchers
(NEDA) -This study can serve as a foundation for future
- research findings can guide NEDA in crafting policies that researchers who would like to pursue the same field
promote balanced growth across sectors while addressing of study and for those who seek to explore other
the challenges posed by the declining agricultural variables as well.
employment share.
SCOPE AND LIMITATIONS

This study will only focus on the agricultural employment share, GDP per
Capita, Share of Industrial Gross Value added in GDP, Share in GDP of Net Foreign
Direct Investment Inflows, and Share in GDP of credit to the private sector. Additionally,
the study will use time series data with 44 observations for the period 1980 to 2023.
The data sources for this research would be secondary sources, specifically coming
from the World Bank, Trading Economics, and STATISTA.
Likewise, the research findings are only limited to the Philippines and may be
irrelevant to other areas with different economic and social factors. This research does
not include the perspectives of policymakers and workers outside the agricultural sector.
REVIEW OF RELATED LITERATURE
Author and Year Country Topic Methodology
Cerutti and Li (2021) Philippines factors influencing labor adjustments in and out of Panel Data Analysis
agriculture
Adam et al. (2024) GCC Countries examines the impact of per capita income, GDP Generalized Method of
growth, foreign direct investment, sectoral Moments (GMM) and
composition, and domestic credit on employment Ordinary Least Squares (OLS)
patterns
Chaudhuri and Developing consequences of Foreign Direct Investment in three-sector general
Banerjee (2010) Countries agricultural land within developing economies equilibrium model and
incorporated it with the Harris
Todaro Hypothesis
Cheong et al. (2023) Vietnam analyzed the employment change structure in Comparative Analysis
Vietnam compared to other Asian countries during
the period of economic reform and industrialization

Oluyemisi and Salihu Nigeria focusing on the relationship between commercial ARDL and VAR
(2021) loans and the agricultural sector
THEORETICAL FRAMEWORK

➢ LEWIS MODEL OR DUAL SECTOR MODEL

The Dual Sector Model, developed by Sir Arthur Lewis in 1954, explains the process of
economic development characterized by a surplus of labor in agriculture. The model shows a shift
from a traditional agricultural economy to a modern industrialized one. This highlights a
change in the labor movement from a low-productivity agricultural sector to a higher-productivity
industrial sector.
From a simplistic viewpoint, it explains that in the early stages, a country has a large surplus
of labor in agriculture, meaning many workers are engaged in low-productivity farming. As the
industrial and service sector grows, it absorbs the workers, offering higher wages and
better opportunities. This shift leads to economic growth and modernization as the economy’s
reliance on agriculture declines (Gollin, 2014).
CONCEPTUAL FRAMEWORK
VARIABLES AND DATA SOURCES
Variables Description Year Sources
Agricultural Employment proxied by the total percentage of 1980-2023 World Bank, Trading
Share (Y) employment rate in the agricultural sector Economics, and STATISTA
per year (% of employment rate).
GDP per Capita (X₁) the total economic output per person, 1980-2023 World Bank
calculated by dividing the gross domestic
product by the total population.
Share of Industrial Gross refers to the industrial sector's share in the 1980-2023 World Bank
Value Added in GDP (X₂) overall GDP (% of GDP).
Share in GDP of Net Foreign captures the impact of foreign capital 1980-2023 World Bank
Direct Investment Inflows investment relative to national economic
(X₃) output. It is measured as net FDI inflows as a
percentage of GDP (% of GDP).
Share in GDP of credit to the represents the availability of credit to private 1980-2023 World Bank
private sector (X₄) enterprises in relation to GDP (% of GDP).
 STATISTICAL METHOD EMPIRICAL MODEL

Trend Analysis Autoregressive Distribution Lag Model

Time Series Analysis
I. Stationarity Test
- Augmented Dickey-Fuller Test
II. ARDL Bounds Test for Cointegration
Autoregressive Distribution Lag Model
III. Error Correction Model
IV. Diagnostic Checking
• Breusch-Godfrey Test
• Breusch-Pagan Test
• Jarque-Bera Test
• Ramsey RESET
• COSUM and COSUMQ
ESTIMATION PROCEDURE

The trends of Agricultural Employment Share, GDP per Capita, Share of Industrial
Gross Value Added in GDP, Share in GDP of Net Foreign Direct Investment Inflows, and
Share in GDP of credit to the private sector will be generated using Microsoft Excel 2020.
Stata Version 12.0 will be used to run the Augmented Dickey-Fuller Test to ensure the series
is a mix of both I(0) and I(1). For the ARDL Bounds Test, Error Correction Model, and
diagnostic checking, Eviews 12 was utilized.
 TREND ANALYSIS

AGRICULTURAL EMPLOYMENT SHARE

• Agricultural employment in the 60
Philippines declined steadily from about
50
51% in 1980 to 22% in 2023.
• The decline was gradual until the mid- 40

1990s. 30

• It accelerated in the late 1990s to the 20

early 2000s due to globalization and 10

environmental changes.
0
• A brief recovery occurred in 2020 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022

during the COVID-19 pandemic, but the

downward trend resumed in 2021.
 TREND ANALYSIS
• This shows a strong upward trend in
GDP Per Capita
the Philippines' GDP per capita from
4000
1980 to 2023, indicating consistent
3500
economic growth. 3000
• Slow growth in the 1980s and early 2500
1990s due to economic 2000
mismanagement, political instability, and 1500
reliance on foreign loans. 1000
• Recovery in the 1990s was hindered by 500
natural disasters and high poverty rates. 0
• A significant surge occurred after the 1980198219841986198819901992199419961998200020022004200620082010201220142016201820202022

2000s, driven by OFW remittances, the

rise of the BPO industry, and improved
macroeconomic policies.
 TREND ANALYSIS

SHARE OF INDUSTRIAL VALUE ADDED IN GDP

• In the early 1980s, the industrial sector
50
contributed about 42% to the 45
Philippines' GDP. 40

• This share has gradually declined to 35

30
around 28–30% in recent years (NEDA, 25

2022). 20
15
• According to Bajpai (2024), labor and 10
resources have largely transitioned 5
0
from agriculture directly to services, 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022
bypassing the industrial sector.
 TREND ANALYSIS

• FDI inflows were low in the 1980s due

SHARE IN GDP OF FOREIGN DIRECT INVESTMENT INFLOWS
to limited investor confidence and
3.5
economic challenges.
3
• Early 1990s growth was driven by 2.5
Globalization, Trade liberalization, and 2
investment-friendly policies. 1.5
• The late 1990s to 2000s experienced 1
sharp fluctuations due to political 0.5
instability, corruption, infrastructure 0
issues, and external shocks -0.5
1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 20002002 2004 2006 2008 20102012 2014 2016 2018 2020 2022

• From 2010 onwards, FDI grew more

steadily.
 TREND ANALYSIS
• In the early 1980s, the ratio was low due
to a less developed financial sector.
SHARE IN GDP OF CREDIT TO THE PRIVATE SECTOR
• After a mid-1980s decline, recovery began
60
in the late 1980s and early 1990s,
50
supported by financial sector expansion
and improved lending conditions. 40

• A significant rise occurred in the mid- 30

1990s, peaking around 1997 due to
20
economic growth and financial liberalization
(Clarete et al., 2018). 10

• From 2010 onward, reforms boosted 0

1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 2020 2022
private sector credit, with private
investment reaching 11.7% of GDP (Tan,
2012)
• A new peak occurred in the late 2010s
and early 2020s, followed by a slight
decline due to the COVID-19 pandemic.
 TIME SERIES ANALYSIS
I. STATIONARITY TEST
T-statistics p-value
Variables
Augmented Dickey-Fuller Test I(0) I(1) I(0) I(1)
Agricultural Employment
-0.123 -5.508 0.9471 0.0000
Share (Agriemp)
The ARDL Bounds Test for
Cointegration can be conducted, specifically GDP Per Capita (GDPpC)
2.052 -5.263 0.9987 0.0000
when the independent variables are a Share of Industrial Value
-1.148 -7.050 0.6955 0.0000
mixture of I(0) and I(1), while the dependent Added in GDP (Indva)
variable is I(1).
Share in GDP of Net
Foreign Direct Investment -3.240 0.0178
Inflows (FDI)

Share in GDP of Credit to

the Private Sector -0.607 -4.089 0.8694 0.0010
(CreditPriv)
TIME SERIES ANALYSIS
II. ARDL Bounds Test

Test Statistics 1% Critical Level 5% Critical Level 10% Critical Level

F-statistics I(0) I(1) I(0) I(1) I(0) I(1)

8.6743 3.81 4.92 3.05 3.97 2.68 3.53

This result confirms the presence of a long-run cointegrating

relationship among the variables. This is supported by the fact that the F-
statistics of 8.6743 exceeds the upper critical values or the I(1) at the 1%
level with a value of 4.92, at the 5% level with 3.97, and at the 10% level
with 3.53.
 TIME SERIES ANALYSIS
Autoregressive Distribution Lag Model

• A 1-unit increase in industrial value-

LONG-RUN ESTIMATES added reduces agricultural
employment by 0.84 in the long run.
Variables Coefficient Std. Error p-value This indicates structural transformation,
as industry grows, agriculture loses
GDPPC 0.0001 0.0003 0.6585
labor.
• An increase in private sector credit
INDVA -0.8398 0.2592 0.0033 can also significantly reduce
agricultural employment, likely
FDI -0.3769 0.1935 0.0623 because credit supports non-agricultural
enterprises.
CREDITPRIV -0.0570 0.0191 0.0062 • On the other hand, GDP per Capita and
Foreign Direct Investments do not
@TREND -0.8954 0.0923 influence Agricultural Employment Share
0.000
in the long run.
 TIME SERIES ANALYSIS
Autoregressive Distribution Lag Model
• The ECM coefficient in this study is -0.7709
SHORT-RUN ESTIMATES and significant at a 5% level, indicating that
Variables Coefficient Std. Error p-value deviations from the long-run equilibrium are
corrected at a rate of approximately 77.1%
D(AGRIEMP(-1)) 0.2134 0.1008 0.0439
per year. This suggests a very strong
D(AGRIEMP(-2)) 0.4776 0.1135 0.0003 speed of adjustment, meaning that
D(AGRIEMP(-3)) 0.1978 0.0989 0.0562 agricultural employment could return to the
equilibrium right away after experiencing
D(INDVA) -0.8601 0.1166 0.0000
shocks.
D(INDVA(-1)) -0.3084 0.1136 0.0116 • The variable industrial value-added
(INDVA) is statistically significant, which
D(CREDITPRIV) -0.0340 0.0243 0.1743
implies that industrial growth continues to pull
D(CREDITPRIV(-1)) 0.0927 0.0296 0.0043 labor out of agriculture.
CointEq(-1) -0.7709 0.0979 0.0000 • The lagged values INDVA(-1) and
CREDITPRIV (-1) is also significant.
Constant 67.389 8.6725 0.0000
TIME SERIES ANALYSIS
IV. Diagnostic Checking

Diagnostic Test Test Statistics p-value

Breusch-Godfrey Test 0.7137 (F-stat) 0.4999

Breusch-Pagan Test 0.7330 (F-stat) 0.7165

Jarque-Bera Test 1.6262 0.4435

Ramsey RESET 2.8335 (F-stat) 0.1748

Overall, these results show that the model used is statistically

robust and reliable. It is also well-suited for analyzing the relationship
between the dependent and independent variables.
 TIME SERIES ANALYSIS
IV. Diagnostic Checking

COSUM and COSUMQ

15 1.4
1.2
10
1.0
5 0.8
0.6
0
0.4
-5 0.2
0.0
-10
-0.2
-15 -0.4
98 00 02 04 06 08 10 12 14 16 18 20 22 98 00 02 04 06 08 10 12 14 16 18 20 22

CUSUM 5% Significance CUSUM of Square s 5% Significance

 SUMMARY AND CONCLUSION
Findings from the ARDL Bounds Test confirmed a statistically significant long-run relationship among
the variables, validating the model’s suitability for exploring long-term dynamics. In the long run, industrial value
added and private sector credit significantly reduce agricultural employment, emphasizing the ongoing structural
transformation of the Philippine economy away from agriculture. On the other hand, FDI and GDP per Capita do
not exert a statistically significant influence on agricultural employment share. The time trend variable is highly
significant and negative, reinforcing the evidence of a persistent decline in agricultural employment over the
decades.

The Error Correction Term (ECM) is negative and highly significant, indicating a 77.1% of short-term
imbalances are corrected within one year- a strong speed of adjustment toward the long-run equilibrium.
Diagnostic tests confirm the robustness and reliability of the model, showing no issues of autocorrelation,
heteroscedasticity, or model misspecification. Stability tests (CUSUM and CUSUMQ) also confirmed that the
model remains stable over time.

In conclusion, the decline in agricultural employment in the Philippines reflects broader

economic transformations driven by industrial expansion. The findings underscore the structural shift in the
labor force as workers move toward higher-productivity sectors, reshaping the Philippine economy.
 RECOMMENDATIONS
• Since industrial expansion significantly reduces agricultural employment, policymakers should promote
stronger linkages between agriculture and industry. Encouraging agribusiness investments, agro-processing
industries, and value-chain development can create more employment opportunities within the agricultural
sector, reducing excessive labor migration to other industries.

• Given the structural shift from agriculture to industry and services, long-term strategies should be
implemented to ensure agricultural productivity and food security. Investments in modern farming
technologies, climate-resilient agricultural practices, and infrastructure improvements in rural areas can help
sustain agricultural employment while increasing sectoral efficiency.

• The delayed but positive impact of credit to the private sector implies that expanding access to rural credit
and financial services, especially for small farmers and agribusinesses— can promote investment and
employment growth in agriculture. Tailored loan products and financial literacy programs may help improve
utilization of credit in agricultural productivities.

• For future research, exploring more macroeconomic variables that affect agricultural employment will be
beneficial. Examining other econometric techniques is also recommended to contribute additional insights to
this field of study.