REVIEW OF THE INVERSE FARM SIZE-EFFICIENCY RELATIONSHIP IN AFRICA:

METHODOLOGICAL ISSUES AND EMPIRICAL EVIDENCE FROM THREE AFRICAN

COUNTRIES

Milu Muyanga, Chewe Nkonde, Godwin Debrah, T.S. Jayne Agricultural, Food and Resource Economics Department

Michigan State University, USA

Photo: Christiaensen and Demery (2007)

Presented at the at the Annual World Bank Conference on Land and Poverty 2015

March 23-27, 2015, Washington, DC

Background

• Renewed interest in the Inverse Farm Size-

Efficiency Relationship (IR) among

development economists

• Guiding land allocation policies for reduced

poverty:

• Are prevailing land policies promoting national goals of

agricultural productivity, food security and poverty

reduction?

• Especially with…

24

Background

• Guiding land allocation policies for reduced

poverty:

• Especially with…

1. Shrinking FARM SIZES due to mounting

POPULATION pressure

2. Changing farm structure-- rising proportion of

farmland among medium-scale farms in Africa

3. Limited non-farm employment opportunities

4. Persistent high poverty rates and food insecurity

23

Total rural population projections

Source: UN Pop Council, 2013

22

Shrinking farm sizes .2

.4.6

.8h

ecta

res p

er

pe

rso

n

1960 1970 1980 1990 2000 2010year

World Sub-Saharan Africa Zambia

Source: World Development Indicators, World Bank

1960-2010Arable land per capita

22

Background

• Guiding land allocation policies for reduced

poverty:

• Especially with…

1. Shrinking farm sizes due to mounting population

pressure

2. Changing FARM STRUCTURE-- rising

proportion of land among medium-scale

farms

3. Youth BULGE and limited NON-FARM employment

opportunities

21

Table 2: Changes in farm structure among small- and medium-scale farmers in Zambia (2009 - 2012)

Landholding size Category

Number of farms % change

(2001-2012)

% of total farmland

Share of landholding

2001* 2009 2012 2009 2012 cultivated (2012)

0 – 2 ha 638,118 916,787 748,771 17.3% 24.1% 16.2% 91.2%

2 – 5 ha 159,039 366,628 418,544 163.2% 33.8% 31.7% 66.4%

5 – 10 ha 20,832 110,436 165,129 692.6% 20.3% 25.0% 49.5%

10 – 20 ha 2,352 35,898 53,454 2272.7% 12.3% 15.0% 36.7%

20 – 100 ha -- 9,030 13,839 53.3%** 9.5% 12.0% 10.9%

Total 820,341 1,438,779 1,399,737 70.6% 100.0% 100.0%

Source: Ministry of Agriculture Crop Forecast Surveys, 2009, 2012. *2001 figures are land under cultivation. ** Growth rate computed from 2009-2012 only. “na” means not available.

Changing farm structure

21

Background

• Guiding land allocation policies for reduced

poverty:

• Especially with…

1. Shrinking farm sizes due to mounting population

pressure

2. Changing farm structure-- rising proportion of

farmland among medium-scale farms in Africa

3. Looming employment challenge: YOUTH

BULGE and limited NON-FARM

employment opportunities

20

Looming employment challenge in SSA

[0-4]

[5-9]

[10-14]

[15-19]

[20-24]

[25-29]

[30-34]

[35-39]

[40-44]

[45-49]

[50-54]

[55-59]

[60-64]

[65-69]

[70-74]

[75-79]

[80+]

-10% -8% -6% -4% -2% 0% 2% 4% 6% 8% 10%

Age pyramid:rural SSA, 2015 Male

Female

62% < 25 years old

20

Tests of the IR hypothesis take on even greater

policy importance in light of recent studies

questioning the viability and even the

objectives of promoting small-scale agriculture

in Africa

19

Contribution [I]

• Explore the IR hypothesis over a much wider

range of farm sizes - a statistically

representative sample of farms between 1 and

100 hectares

• Inform current policy discussions about how

governments should allocate unutilized/underutilized

land in order to achieve national equity and

productivity goals

• Unutilized/underutilized land is being claimed and

transferred at a very rapid pace in some countries

18

Contribution [II]

• Number of studies have conventionally measured productivity as yield and or net value of crop production per unit area of land

• Our study is based on a wider set of productivity measures: • Net value of total crop production per unit of area

planted (land productivity)

• Net value of crop production per adult labor unit (labor productivity)

• Cost of production per metric ton of maize produced

(cost effectiveness)

17

Contribution [III]

• Account for both variable and fixed

costs when computing the cost of

production.

• Most of the prior studies typically ignored

fixed and labor costs

• Led to overstated productivity of farms

with high fixed and labor costs

16

Data sources• KENYA

• Data on about 300 smallholder (0-5ha) farm households came from Egerton University/Tegemeo Institute collected in 2010 in 5 counties in Western Kenya

• A survey involving 200 medium scale (5-100ha) farmers in the same counties was collected in 2012

• GHANA• Data on 498 small and medium farms came from

four districts in Southern Ghana

• ZAMBIA• Smallholder and medium-scale household surveys

from 6 districts

15

• Descriptive analysis

• Econometric analysis

is the measure of productivity is Area planted in acres is a vector of covariates district dummies is the error term

Methods

14

Results

22

KENYA: Medium-scale farms productivity

13

Gross value

Net value

20

60

10

0

'00

0K

Sh

/ha p

lan

ted

25%50% 75% 95%0 5 10 15 20 25

land -- planted (ha)

Figure 2a: Value of crop production per hectare planted

KENYA: Smallholder farms

productivity

12

Farm income

Crop income

02

04

06

08

01

00

'00

0 K

Sh/h

a p

lante

d

25% 50% 75% 95%0 1 2 3 4 5 6

land -- planted (ha)

Figure 2b: Farm and crop income per hectare planted

KENYA: Smallholder farms

productivity

11

KENYA: Farm production cost2

02

2.5

25

27

.53

03

2.5

35

'00

0K

Sh

/ha p

lan

ted

25%50% 75% 95%0 10 20

land -- planted (ha)

Figure 5a: Total crop production costs per hectare planted

10

KENYA: Farm production costs- components

9

-50

51

01

52

02

53

0

'00

0K

Sh

/ha p

lan

ted

25%50% 75% 95%0 5 10 15 20 25

land -- planted (ha)

Fertilizer Seed

Land preparation Labor

Fixed costs

Figure 6: Total crop production costs by components

KENYA: Labor costs- components

8

GHANA: Farm productivity

7

bivariate

exogenous

exogenous +

24

68

10sh

(net

val

ue o

f cro

p pr

oduc

tion

per a

cre)

in G

h C

edis

0 100 200 300Landholding size (acre)

bivariate

exogenous

exogenous +

-50

510

sh (n

et v

alue

of c

rop

prod

uctio

n pe

r acr

e) in

Gh

Cedi

s

0 100 200 300area planted(acres)

Net value of production on Landholdings in Acres

Net value of production on Area planted in Acres

GHANA: Farm productivity

6

Family labor productivity on Landholdings in Acres

Family labor productivity on Area planted in Acres

bivariate

exogenous

exogenous +

910

1112

1314

ln(fa

mily

labo

r pro

duct

ivity

) in

Gh

Cedi

s

0 100 200 300Landholding size (acre)

bivariate

exogenous

exogenous +

810

1214

ln(fa

mily

labo

r pro

duct

ivity

) in

Gh

Cedi

s

0 100 200 300area planted (acre)

GHANA: Farm productivity

5

Cost of maize production on Landholdings in Acres

Cost of maize production on Area planted in Acres

bivariate

exogenous

exogenous +

45

67

8ln

(cos

t per

met

ric to

n of

pro

ducin

g m

aize

) in

Gh C

edis

0 100 200 300Landholding size (acre)

bivariateexogenous

exogenous +

02

46

8ln

(cos

t per

met

ric to

n of

pro

ducin

g m

aize

) in

Gh C

edis

0 100 200 300area planted(acre)

ZAMBIA: Farm productivity & efficiency

4

Policy implications

Large scale

Medium-scale

(5-100 ha)

Small-scale

(0-5 ha)

Total land controlled

Potentially available cropland

remaining

Millions of hectares

Ghana 3.08 4.21 5.08 = 12.37 3.56

Kenya 0.69 0.84 2.63 = 4.16 1.01

Zambia 2.11 2.47 2.09 = 6.67 3.35

To which scale of farming should the remaining PAC be allocated?

3

Source: Jayne et al. 2014 (JIA)

Policy implications

1. Production efficiency, while relevant, should not

be the ONLY factor in guiding agricultural and land

policies

• Which scale has the largest multiplier and employment

effects?

• Which scale has the highest marginal propensity to

consume?

2. All depends on the government’s development

objective:

• Production for domestic food self sufficiency and export

market?

• Broad based growth for reduced food insecurity and poverty

reduction?

3. In in all, the changing farm structure is going to

continue in the next 5-10 years

• Drivers: political economy factors and market forces

2

Policy implications

1. Production efficiency, while relevant, should not

be the ONLY factor in guiding agricultural and land

policies

• Which scale has the largest multiplier and employment

effects?

• Which scale has the highest marginal propensity to

consume?

2. All depends on the government’s development

objective:

• Production for domestic food self sufficiency and export

market?

• Broad based growth for reduced food insecurity and poverty

reduction?

3. In in all, the changing farm structure is going to

continue in the next 5-10 years

• Drivers: political economy factors and market forces

2

Policy implications

1. Production efficiency, while relevant, should not

be the ONLY factor in guiding agricultural and land

policies

• Which scale has the largest multiplier and employment

effects?

• Which scale has the highest marginal propensity to

consume?

2. All depends on the government’s development

objective:

• Production for domestic food self sufficiency and export

market?

• Broad based growth for reduced food insecurity and poverty

reduction?

3. In all, the changing farm structure is going to

continue in the next 5-10 years

• Drivers: political economy factors and market forces

2

1

Conclusion

• Land policies will determine whether millions of

rural Africans will make a decent livelihood

• How supportive the land allocation and agricultural policies

are to smallholders

• African leaders may soon realize that political

stability will depend on how the remaining land is

distributed and the profitability of family farming

Acknowledgements