Niger

Notes:

  • 1 Int’l $ = 257.60 XOF (CFA) using 2020 World Bank PPP conversion rates (1 Int’l $ = 1 USD)
  • Focus crop = cowpea and millet
  • All costs are reported per hectare. Inspection, certification and other marketing costs are assumed for the entire farm. Labeling and packaging are per kg.
  • Some farmers grow multiple crops

Survey Recodes

xrate <- 257.60

# Load respondent data
hh <- fread("../data/nre/hh.csv")
group <- fread("../data/nre/group.csv")

There are 24 variables and 80 observations in this set. A summary is shown below.

print(dfSummary(hh), max.tbl.height=500)

Error : thematic doesn’t (yet) support the ‘PNG’ graphics device. Please report this error to https://github.com/rstudio/thematic/issues/new

Variable Stats / Values Freqs (% of Valid) Valid Missing
Code
[character]		 1. Chadakori_MI_FG1
2. Chadakori_MI_FG2
3. Chadakori_NI_FG1
4. Chadakori_NI_FG2
5. Sabon_Machi_MI_FG1
6. Sabon_Machi_MI_FG2
7. Sabon_Machi_NI_FG1
8. Sabon_Machi_NI_FG2		 10 (12.5%)
10 (12.5%)
10 (12.5%)
10 (12.5%)
10 (12.5%)
10 (12.5%)
10 (12.5%)
10 (12.5%)		 80
(100.0%)		 0
(0.0%)
Group
[character]		 1. Boukassa_Noma
2. Cigaban_Matta
3. HIMMA
4. Yada_Iri		 10 (12.5%)
10 (12.5%)
40 (50.0%)
20 (25.0%)		 80
(100.0%)		 0
(0.0%)
Crop
[character]		 1. cowpea
2. millet		 40 (50.0%)
40 (50.0%)		 80
(100.0%)		 0
(0.0%)
District
[character]		 1. Dakoro
2. Guidan Roumdji		 40 (50.0%)
40 (50.0%)		 80
(100.0%)		 0
(0.0%)
Location
[character]		 1. Chadakori
2. Sabon Machi		 40 (50.0%)
40 (50.0%)		 80
(100.0%)		 0
(0.0%)
Date
[IDate, Date]		 1. 2021-04-28
2. 2021-04-30		 40 (50.0%)
40 (50.0%)		 80
(100.0%)		 0
(0.0%)
Age
[character]		 1. 15-29
2. 30+		 9 (11.2%)
71 (88.8%)		 80
(100.0%)		 0
(0.0%)
Sex
[character]		 1. female
2. male		 32 (40.0%)
48 (60.0%)		 80
(100.0%)		 0
(0.0%)
Cost seed per ha (LCU)
[integer]		 Mean (sd) : 7136.9 (5225.8)
min < med < max:
1200 < 5000 < 25000
IQR (CV) : 6000 (0.7)		 30 distinct values 80
(100.0%)		 0
(0.0%)
Cost of fertilizer per ha (LCU)
[integer]		 Mean (sd) : 27441.9 (14019.1)
min < med < max:
0 < 31250 < 75000
IQR (CV) : 18000 (0.5)		 35 distinct values 80
(100.0%)		 0
(0.0%)
Cost of pesticide per ha (LCU)
[integer]		 Mean (sd) : 6648.1 (8051.2)
min < med < max:
0 < 3000 < 36000
IQR (CV) : 9500 (1.2)		 29 distinct values 80
(100.0%)		 0
(0.0%)
Cost of transport per ha (LCU)
[integer]		 Mean (sd) : 5822.8 (4742.9)
min < med < max:
0 < 5000 < 25000
IQR (CV) : 3187.5 (0.8)		 25 distinct values 80
(100.0%)		 0
(0.0%)
Labor cost (LCU)
[integer]		 Mean (sd) : 35362.5 (20947.8)
min < med < max:
0 < 30000 < 95000
IQR (CV) : 28500 (0.6)		 41 distinct values 80
(100.0%)		 0
(0.0%)
Inspection / certification Fees (LCU)
[integer]		 Mean (sd) : 2031.2 (2028.8)
min < med < max:
0 < 2000 < 10000
IQR (CV) : 3000 (1)		 0 : 31 (38.8%)
500 : 1 ( 1.2%)
1000 : 1 ( 1.2%)
1500 : 4 ( 5.0%)
2000 : 5 ( 6.2%)
3000 : 25 (31.2%)
5000 : 12 (15.0%)
10000 : 1 ( 1.2%)		 80
(100.0%)		 0
(0.0%)
Labelling costs per kg (LCU)
[integer]		 Mean (sd) : 123.9 (232.1)
min < med < max:
0 < 0 < 1400
IQR (CV) : 131.2 (1.9)		 13 distinct values 80
(100.0%)		 0
(0.0%)
Packaging costs per kg (LCU)
[integer]		 Mean (sd) : 98.2 (100.3)
min < med < max:
0 < 100 < 500
IQR (CV) : 112 (1)		 22 distinct values 80
(100.0%)		 0
(0.0%)
Other marketing costs? (LCU)
[integer]		 Mean (sd) : 931.2 (3474.8)
min < med < max:
0 < 0 < 30000
IQR (CV) : 1000 (3.7)		 0 : 58 (72.5%)
1000 : 6 ( 7.5%)
1500 : 4 ( 5.0%)
2000 : 4 ( 5.0%)
2500 : 1 ( 1.2%)
3000 : 4 ( 5.0%)
5000 : 2 ( 2.5%)
30000 : 1 ( 1.2%)		 80
(100.0%)		 0
(0.0%)
Estimated Yield (kg/ha)
[integer]		 Mean (sd) : 684.6 (280)
min < med < max:
100 < 700 < 1500
IQR (CV) : 300 (0.4)		 19 distinct values 80
(100.0%)		 0
(0.0%)
Selling price of seed per kg (LCU)
[integer]		 Mean (sd) : 825 (354.2)
min < med < max:
200 < 800 < 1500
IQR (CV) : 500 (0.4)		 200 : 1 ( 1.2%)
250 : 4 ( 5.0%)
300 : 1 ( 1.2%)
350 : 1 ( 1.2%)
500 : 21 (26.2%)
750 : 11 (13.8%)
800 : 3 ( 3.8%)
1000 : 28 (35.0%)
1500 : 10 (12.5%)		 80
(100.0%)		 0
(0.0%)
Selling price of grain per kg (LCU) at sowing
[integer]		 Mean (sd) : 419.4 (216.8)
min < med < max:
200 < 362.5 < 1200
IQR (CV) : 250 (0.5)		 16 distinct values 80
(100.0%)		 0
(0.0%)
Selling price of grain per kg (LCU) at harvest
[integer]		 Mean (sd) : 245.9 (189.5)
min < med < max:
100 < 150 < 1000
IQR (CV) : 167 (0.8)		 17 distinct values 80
(100.0%)		 0
(0.0%)
How many kg were sold in the season?
[integer]		 Mean (sd) : 353.4 (289.8)
min < med < max:
0 < 300 < 1500
IQR (CV) : 385 (0.8)		 22 distinct values 80
(100.0%)		 0
(0.0%)
What was your expected gross margin?
[integer]		 Mean (sd) : 137489.4 (145046.5)
min < med < max:
50 < 73500 < 5e+05
IQR (CV) : 180000 (1.1)		 36 distinct values 80
(100.0%)		 0
(0.0%)
How long have you been a member of this group?
[integer]		 Mean (sd) : 11.9 (7.9)
min < med < max:
1 < 11 < 23
IQR (CV) : 15 (0.7)		 18 distinct values 80
(100.0%)		 0
(0.0%)

Recode variable names.

setnames(hh, lbl$label, lbl$code, skip_absent=T)

Additional recodes for categorical variables.

setorder(hh, adm2_nm, adm3_nm, group, gender, crop)

hh[, `:=`(
  hhid = paste("NRE", gsub(" ", "0", format(1:.N, width=3)), sep=""),
  iso3 = "NRE",
  crop = factor(crop),
  adm2_nm = factor(adm2_nm),
  adm3_nm = factor(adm3_nm),
  group = factor(group, levels=c(
    "Boukassa_Noma",
    "Cigaban_Matta",
    "Yada_Iri",
    "HIMMA"
  ), labels=c(
    "Boukassa Noma",
    "Cigaban Matta",
    "Yada Iri",
    "Himma"    
  )),
  gender = factor(gender, levels=c("male", "female"), labels=c("Male", "Female")),
  age = factor(age, levels=c("15-29", "30+"), labels=c("< 30", "≥ 30")),
  years = factor(member_years >= 5, levels=c(F, T), labels=c("< 5", "≥ 5"))  
)]

Constructed Variables

Farmers report both expected yields yield_ha_kg and actual sales in the last season sales_ha_kg, so we can construct both expected and realized costs in monetary terms costs_exp_ha_lcu and costs_real_ha_lcu. Note that we then use realized sales to calculate profitability metrics. We assume that in Niger marketing costs for both cowpea and millet are reported for the entire farm (not by kg as in Vietnam).

hh[, 
  tran_ha_lcu := as.numeric(tran_ha_lcu)
][, 
  tran_ha_lcu := fifelse(is.na(tran_ha_lcu), 0, tran_ha_lcu)
][, `:=`(
  # Expected costs
  costs_exp_ha_lcu = 
    # Per ha costs
    seed_ha_lcu + fert_ha_lcu + pest_ha_lcu + tran_ha_lcu + labor_ha_lcu + 
    cert_lcu + mark_kg_lcu +
    # Per kg costs
    yield_ha_kg * (labl_kg_lcu + pckg_kg_lcu),  
  
  # Realized costs
  costs_real_ha_lcu = 
    # Per ha costs
    seed_ha_lcu + fert_ha_lcu + pest_ha_lcu + tran_ha_lcu + labor_ha_lcu + 
    cert_lcu + mark_kg_lcu +
    # Per kg costs
    sales_ha_kg * (labl_kg_lcu + pckg_kg_lcu)
)]

hh[, summary(costs_exp_ha_lcu)]
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   21100   90719  172200  242594  299875  896000
hh[, summary(costs_real_ha_lcu)]
##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##   21100   71775  126550  161174  186500  729500

Using realized costs and sales, we construct gross margin per ha margin_ha_lcu, total sales sales_ha_sh and profit margin margin_ha_sh per unit of (variable) input costs, and costs_ha_ppp, sales_ha_ppp and margin_ha_ppp in PPP terms to allow for comparisons across groups and countries.

We also construct a measure of total factor productivity tfp as expected output per unit of (expected) input costs. Strictly speaking it is only “partial factor productivity” here because we don’t include the rental cost of land, land preparation costs, irrigation costs, and the costs of animal and mechanical implements.

hh[, `:=`(
  sales_exp_ha_lcu = yield_ha_kg * sales_kg_lcu,
  sales_real_ha_lcu = sales_ha_kg * sales_kg_lcu
)][, `:=`(
  margin_ha_lcu = sales_real_ha_lcu - costs_real_ha_lcu
)][, `:=`(
  # Shares
  sales_ha_sh = sales_real_ha_lcu / costs_real_ha_lcu,
  margin_ha_sh = margin_ha_lcu / costs_real_ha_lcu,
  # PPP$
  costs_ha_ppp = costs_real_ha_lcu / xrate,
  sales_ha_ppp = sales_real_ha_lcu / xrate,
  margin_ha_ppp = margin_ha_lcu / xrate
)][, `:=`(
  tfp = yield_ha_kg / (costs_exp_ha_lcu / xrate)
)]

Note that 6 respondents indicate having no sales (and hence a negative gross margin) this past season (crop failure or decision not to sell?).

tmp <- hh[sales_ha_kg==0, .(
  hhid, code, group, crop, 
  sales_exp_ha_lcu, sales_real_ha_lcu, margin_ha_lcu, yield_ha_kg)]

kbl(
  setnames(tmp, lbl[data.table(code=names(tmp)), on=.(code), label]),
  format.args=list(big.mark=","))
Unique farmer/crop ID Code Group Crop Gross sales Gross sales per ha (realized yield) Gross margin per ha Estimated Yield (kg/ha)
NRE042 Chadakori_MI_FG1 Boukassa Noma millet 375,000 0 -28,200 1,500
NRE046 Chadakori_MI_FG1 Boukassa Noma millet 300,000 0 -88,250 1,000
NRE047 Chadakori_MI_FG1 Boukassa Noma millet 25,000 0 -21,900 100
NRE056 Chadakori_MI_FG2 Cigaban Matta millet 250,000 0 -75,000 500
NRE074 Chadakori_NI_FG2 Yada Iri cowpea 400,000 0 -94,000 800
NRE076 Chadakori_NI_FG2 Yada Iri cowpea 1,200,000 0 -134,000 800

Below we append some of the information that was recorded at the group level.

# Same recodes in the group-level dataset
group[, Group := factor(Group, levels=c(
  "Boukassa Noma",
  "Cigaban Matta",
  "Yada Iri",
  "HIMMA"
), labels=c(
  "Boukassa Noma",
  "Cigaban Matta",
  "Yada Iri",
  "Himma"    
))]

# Merge
hh[group, on=.(group=Group), `:=`(
  group_year = `Established`,
  group_size = `Members`,
  seasons = `Seasons`,
  irrigated = `Irrigation`,
  market_access = `Market access`,
  ttrade = `Transboundary trade`
)]

kbl(group, align="lllllc")
Group Region Department Commune Location Market access Irrigation Seasons Transboundary trade Members Established
Boukassa Noma Maradi Guidan Roumdji Chadakori Kouroungoussaou – – 1 Yes 56 1998
Yada Iri Maradi Guidan Roumdji Chadakori Chadakori – – 1 No 80 1998
Cigaban Matta Maradi Guidan Roumdji Chadakori Chadakori – – 1 No 33 1998
Himma Maradi Dakoro Sabon Machi Sabon Machi – – 1 Yes 44 1999

Finally we normalize all cost line items into a “long” table hh_prod_cost for charting. Again we use farmers’ realized costs here.

# Normalize production cost table per ha
hh_prod_cost <- hh[, .(hhid,
  Seeds = seed_ha_lcu, 
  Fertilizer = fert_ha_lcu, 
  Pesticides = pest_ha_lcu, 
  Labor = labor_ha_lcu, 
  Transport = tran_ha_lcu, 
  Certification = cert_lcu,
  Labeling = sales_ha_kg * labl_kg_lcu,
  Packaging = sales_ha_kg * pckg_kg_lcu,
  Marketing = mark_kg_lcu
)]

hh_prod_cost <- melt(hh_prod_cost, id.vars=1, value.name="lcu", variable.name="type")

And we lump all marketing costs into a single category.

levels(hh_prod_cost$type) <- levels(hh_prod_cost$type)[c(1,2,3,4,9,9,9,9,9)]

hh_prod_cost <- hh_prod_cost[, .(
  lcu = sum(lcu, na.rm=T)
), by=.(hhid, type)
][, `:=`(
  # Add cost shares and PPP terms
  share = lcu/sum(lcu, na.rm=T),
  ppp = lcu/xrate
), by=.(hhid)
][hh, on=.(hhid), `:=`(
  # Add classes
  group = group,
  gender = gender,
  age = age,
  years = years,  
  crop = crop
)]

Descriptive Statistics

Respondent Characteristics

Breakdown by categorical variables. Note that farmers in HIMMA grow both crops.

ggplot(
  hh[, .N, by=.(group, age, gender, crop, years)],
  aes(axis1=crop, axis2=gender, axis3=age, axis4=years, y=N)) +
  geom_alluvium(aes(fill=group), width=1/4, alpha=.7, color="white") +
  geom_stratum(width=1/4) +
  geom_text(stat="stratum", aes(label=after_stat(stratum)), angle=90, size=2.2) +
  scale_x_discrete(limits=c("Crop", "Gender", "Age", "Years in Seed Club")) +
  labs(y=NULL, fill="Seed Club",
    title = "Categories of Survey Respondents - Niger",
    subtitle = "Stratified by seed club") +
  theme_def(axis.text=element_text(face="bold"))

Showing contingency table between each pair of categorical variables (seed club group, gender, age age, and years in seed club years).

ttt_ftable(hh, vars=c("group", "gender", "years"))
Contingency Table (% of respondents)
group gender < 5 ≥ 5 Sum

N = 80 | Mantel-Haenszel chi-squared = 6.74 | p-value = 0.0806
Boukassa Noma Male 3.8 3.8 7.5
Female 2.5 2.5 5
Sum 6.2 6.2 12.5
Cigaban Matta Male 3.8 6.2 10
Female 1.2 1.2 2.5
Sum 5 7.5 12.5
Yada Iri Male 2.5 16.2 18.8
Female 1.2 5 6.2
Sum 3.8 21.3 25
Himma Male 1.2 22.5 23.8
Female 5 21.2 26.2
Sum 6.2 43.8 50
Sum Male 11.2 48.8 60
Female 10 30 40
Sum 21.3 78.8 100

Seed Production Costs

General breakdown and distribution of input costs across seed clubs, gender, years in seed club, and input type.

ttt(costs_ha_ppp ~ group | gender+years, data=hh, render=fmt,
  caption="Total Input Costs in Absolute Terms (PPP$ / ha) - Niger")
Total Input Costs in Absolute Terms (PPP$ / ha) - Niger
group Statistic < 5 ≥ 5
Male Female Male Female
Boukassa Noma mean 1,112 1,102 1,177 674
median 901 1,102 613 674
sd 894 1,414 1,458 798
Cigaban Matta mean 219 250 322 149
median 231 250 291 149
sd 30 NA 107 NA
Yada Iri mean 356 404 724 628
median 356 404 599 481
sd 209 NA 547 309
Himma mean 696 402 806 418
median 696 379 728 307
sd NA 262 356 342

Boxplots with mean comparison p-value and significance levels. When more than two levels, each level is compared to the group mean.

(ns : p > 0.05, * : p ≤ 0.05, ** : p ≤ 0.01, *** = p ≤ 0.001, **** = p ≤ 0.0001)

ggBoxTest(hh, 
  aes(crop, costs_ha_ppp, color=gender, fill=gender), 
  grp.c=aes(group=crop), grp.s=aes(group=gender)) +
  scale_y_continuous(labels=comma) +
  labs(x="", y="", fill="", color="",
    title="Total Input Costs (PPP$ / ha) - Niger",
    subtitle="Stratified by crop and gender") +
  theme_def(legend.position="top")

ggBoxTest(hh, 
  aes(crop, costs_ha_ppp, color=group, fill=group),
  grp.c=aes(group=crop), grp.s=aes(group=group)) +
  scale_y_continuous(labels=comma) + 
  labs(x="", y="", fill="", color="",
    title="Total Input Costs (PPP$ / ha) - Niger",
    subtitle="Stratified by crop") +
  theme_def(legend.position="top")

Breakdown across categories of farm input.

ttt(ppp ~ type | gender+crop, data=hh_prod_cost, render=fmt,
  caption="Input Costs in Absolute Terms by Gender (PPP$ / ha) - Niger")
Input Costs in Absolute Terms by Gender (PPP$ / ha) - Niger
type Statistic cowpea millet
Male Female Male Female
Seeds mean 35 36 26 14
median 35 23 17 12
sd 18 24 24 7
Fertilizer mean 113 112 132 71
median 136 140 130 68
sd 40 40 68 51
Pesticides mean 50 41 5 4
median 47 27 4 2
sd 32 32 4 5
Labor mean 185 152 108 90
median 155 116 103 78
sd 87 87 55 50
Marketing mean 349 250 426 253
median 218 148 31 41
sd 385 251 686 501 
tbl <- hh_prod_cost[, .(
  ppp = mean(ppp, na.rm=T)
), keyby=.(gender, crop, type)]

ggplot(tbl, aes(gender, ppp, fill=type)) +
  geom_bar(stat="identity", position="fill", alpha=.7, width=.6, color="white") +
  scale_y_continuous(labels=percent) +
  facet_wrap(~crop) +
  labs(y="", x="",
    title="Breakdown of Input Costs by Category - Niger",
    subtitle="Stratified by crop and gender") +
  theme_def(legend.position="right")

ttt(ppp ~ type | years+crop, data=hh_prod_cost, render=fmt,
  caption="Input Costs in Absolute Terms by Years in Seed Group (PPP$ / ha) - Niger")
Input Costs in Absolute Terms by Years in Seed Group (PPP$ / ha) - Niger
type Statistic cowpea millet
< 5 ≥ 5 < 5 ≥ 5
Seeds mean 25 37 12 23
median 22 35 12 13
sd 12 20 4 21
Fertilizer mean 92 116 90 104
median 97 140 70 116
sd 55 36 64 68
Pesticides mean 38 49 5 4
median 29 49 4 3
sd 28 33 4 5
Labor mean 147 181 94 100
median 164 155 82 78
sd 86 87 49 55
Marketing mean 205 342 418 303
median 114 198 23 76
sd 195 372 764 530 
tbl <- hh_prod_cost[, .(
  ppp = mean(ppp, na.rm=T)
), keyby=.(years, crop, type)]

ggplot(tbl, aes(years, ppp, fill=type)) +
  geom_bar(stat="identity", position="fill", alpha=.7, width=.6, color="white") +
  scale_y_continuous(labels=percent) +
  facet_wrap(~crop) +
  labs(y="", x="",
    title="Breakdown of Input Costs by Category - Niger",
    subtitle="Stratified by crop and years in seed club") +
  theme_def(legend.position="right")

ttt(ppp ~ type | group+crop, data=hh_prod_cost, render=fmt,
  caption="Input Costs in Absolute Terms by Seed Group (PPP$ / ha) - Niger")
Input Costs in Absolute Terms by Seed Group (PPP$ / ha) - Niger
type Statistic cowpea millet
Yada Iri Himma Boukassa Noma Cigaban Matta Himma
Seeds mean 32 40 20 13 23
median 31 35 12 12 12
sd 13 24 18 6 22
Fertilizer mean 124 101 101 118 91
median 140 110 66 125 107
sd 33 43 103 47 52
Pesticides mean 29 65 4 5 4
median 21 64 4 4 3
sd 20 32 1 6 5
Labor mean 147 205 104 85 102
median 150 204 96 71 78
sd 69 94 50 41 59
Marketing mean 321 323 812 45 241
median 176 246 433 23 82
sd 435 256 935 70 369 
tbl <- hh_prod_cost[, .(
  ppp = mean(ppp, na.rm=T)
), keyby=.(group, crop, type)]

ggplot(tbl, aes(group, ppp, fill=type)) +
  geom_bar(stat="identity", position="fill", alpha=.7, width=.6, color="white") +
  scale_y_continuous(labels=percent) +
  facet_wrap(~crop, scales="free_x") +
  labs(y="", x="",
    title="Breakdown of Input Costs by Category - Niger",
    subtitle="Stratified by crop and seed club") +
  theme_def(legend.position="right")

Are there significant differences across groups? We first compare input cost shares across gender, then across seed clubs.

ggBoxTest(hh_prod_cost, 
  aes(type, share, color=gender, fill=gender),
  grp.c=aes(group=type), grp.s=aes(group=gender)) +
  scale_y_continuous(labels=percent) +
  facet_wrap(~crop) +  
  labs(x="", y="", fill="", color="",
    title="Input Costs by Category (Percent of Total Costs by Ha) - Niger",
    subtitle="Stratified by crop and gender") +
  theme_def(legend.position="top")

ggBoxTest(hh_prod_cost, 
  aes(type, share, color=years, fill=years),
  grp.c=aes(group=type), grp.s=aes(group=years)) +
  scale_y_continuous(labels=percent) +
  facet_wrap(~crop) +  
  labs(x="", y="", fill="", color="",
    title="Input Costs by Category (Percent of Total Costs by Ha) - Niger",
    subtitle="Stratified by crop and years in seed club") +
  theme_def(legend.position="top")

ggBoxTest(hh_prod_cost, 
  aes(type, share, color=group, fill=group),
  grp.c=aes(group=type), grp.s=aes(group=group)) +
  scale_y_continuous(labels=percent) +
  facet_wrap(~crop) +  
  labs(x="", y="", fill="", color="",
    title="Input Costs by Category (PPP$ by Hectare) - Niger",
    subtitle="Stratified by crop and seed club") +
  theme_def(legend.position="top")

Differences between the 2 crops.

ggBoxTest(hh_prod_cost, 
  aes(type, share, color=crop, fill=crop),
  grp.c=aes(group=type), grp.s=aes(group=crop)) +
  scale_y_continuous(labels=percent) +
  labs(x="", y="", fill="", color="",
    title="Input Costs by Category (PPP$ by Hectare) - Niger",
    subtitle="Stratified by crop") +
  theme_def(legend.position="top")

Efficiency

Differences in productivity measures (expected seed yields and actual sales) across groups.

ttt(yield_ha_kg ~ group | gender+crop, data=hh, render=fmt,
  caption="Expected Seed Yield (kg / ha) - Niger")
Expected Seed Yield (kg / ha) - Niger
group Statistic cowpea millet
Male Female Male Female
Boukassa Noma mean NA NA 850 838
median NA NA 900 850
sd NA NA 459 565
Cigaban Matta mean NA NA 688 340
median NA NA 700 340
sd NA NA 236 198
Yada Iri mean 562 620 NA NA
median 800 600 NA NA
sd 279 179 NA NA
Himma mean 693 583 990 697
median 700 600 1,000 700
sd 169 214 256 170 
ttt(sales_ha_kg ~ group | gender+crop, data=hh, render=fmt,
  caption="Realized Seed Sales (kg / ha) - Niger")
Realized Seed Sales (kg / ha) - Niger
group Statistic cowpea millet
Male Female Male Female
Boukassa Noma mean NA NA 433 251
median NA NA 400 252
sd NA NA 427 288
Cigaban Matta mean NA NA 319 170
median NA NA 300 170
sd NA NA 181 99
Yada Iri mean 351 280 NA NA
median 250 300 NA NA
sd 389 110 NA NA
Himma mean 536 334 293 276
median 550 250 250 300
sd 265 355 209 205 
ttt(yield_ha_kg ~ group | years+crop, data=hh, render=fmt,
  caption="Realized Seed Sales (kg / ha) - Niger")
Realized Seed Sales (kg / ha) - Niger
group Statistic cowpea millet
< 5 ≥ 5 < 5 ≥ 5
Boukassa Noma mean NA NA 670 1,020
median NA NA 700 1,000
sd NA NA 315 572
Cigaban Matta mean NA NA 620 617
median NA NA 650 600
sd NA NA 105 343
Yada Iri mean 467 596 NA NA
median 400 800 NA NA
sd 306 250 NA NA
Himma mean 433 700 850 762
median 400 700 850 700
sd 58 170 71 239 
ttt(sales_ha_kg ~ group | years+crop, data=hh, render=fmt,
  caption="Realized Seed Sales (kg / ha) - Niger")
Realized Seed Sales (kg / ha) - Niger
group Statistic cowpea millet
< 5 ≥ 5 < 5 ≥ 5
Boukassa Noma mean NA NA 381 340
median NA NA 500 200
sd NA NA 363 422
Cigaban Matta mean NA NA 298 283
median NA NA 300 250
sd NA NA 45 232
Yada Iri mean 167 363 NA NA
median 100 300 NA NA
sd 115 359 NA NA
Himma mean 367 494 300 278
median 400 500 300 275
sd 58 322 141 209

Differences in efficiency measures across gender with mean comparison (Wilcoxon) p-value.

ggBoxTest(hh, aes(gender, yield_ha_kg, color=gender, fill=gender), cp=list(1:2)) +
  scale_y_continuous(labels=comma) +
  facet_wrap(~crop, scales="free_y") +  
  labs(x="", y="", fill="",
    title="Expected Seed Yield (kg / ha) - Niger",
    subtitle="Stratified by crop and gender") +
  theme_def(legend.position="none")

ggBoxTest(hh, aes(gender, sales_ha_ppp, color=gender, fill=gender), cp=list(1:2)) +
  scale_y_continuous(labels=comma) +
  facet_wrap(~crop, scales="free_y") +  
  labs(x="", y="", fill="",
    title="Total Seed Sales (PPP$ / ha) - Niger",
    subtitle="Stratified by gender") +
  theme_def(legend.position="none")

Differences in efficiency measures by years in seed club with mean comparison (Wilcoxon) p-value.

ggBoxTest(hh, aes(years, yield_ha_kg, color=years, fill=years), cp=list(1:2)) +
  scale_y_continuous(labels=comma) +
  facet_wrap(~crop, scales="free_y") +  
  labs(x="", y="", fill="",
    title="Expected Seed Yield (kg / ha) - Niger",
    subtitle="Stratified crop and by years in seed club") +
  theme_def(legend.position="none")

ggBoxTest(hh, aes(years, sales_ha_ppp, color=years, fill=years), cp=list(1:2)) +
  scale_y_continuous(labels=comma) +
  facet_wrap(~crop, scales="free_y") +  
  labs(x="", y="", fill="",
    title="Total Seed Sales (PPP$ / ha) - Niger",
    subtitle="Stratified crop and by years in seed club") +
  theme_def(legend.position="none")

Differences in efficiency measures across seed clubs with global ANOVA p-value.

ggBoxTest(hh, aes(group, yield_ha_kg, color=group, fill=group)) +
  scale_x_discrete(labels=label_wrap(5)) +
  scale_y_continuous(labels=comma) +
  facet_wrap(~crop, scales="free_y") +
  labs(x="", y="", fill="",
    title="Expected Seed Yield (Kg / ha) - Niger",
    subtitle="Stratified by crop and seed club") +
  theme_def(legend.position="none")

ggBoxTest(hh, aes(group, sales_ha_ppp, color=group, fill=group)) +
  scale_x_discrete(labels=label_wrap(5)) +
  scale_y_continuous(labels=comma) +
  facet_wrap(~crop, scales="free_y") +
  labs(x="", y="", fill="",
    title="Total Seed Sales (PPP$ / ha) - Niger",
    subtitle="Stratified crop and by seed club") +
  theme_def(legend.position="none")

Looking at production frontiers (units of output vs. units of input). We expect S-shape curves with farmers at different levels of technical efficiency along the curve.

Note that in the approximated curves below we remove outliers with total input costs over PPP$ 2,000 per ha.

outlier <- hh[costs_ha_ppp > mean(costs_ha_ppp) + 3*sd(costs_ha_ppp), hhid]

kbl(
  caption="Farmers with total input costs > mean + 3*sd",
  hh[hhid %in% outlier, .(hhid, group, crop, yield_ha_kg, costs_ha_ppp)],
  format.args=list(big.mark=",", digits=0))
Tab. 3: Farmers with total input costs > mean + 3*sd
hhid group crop yield_ha_kg costs_ha_ppp
NRE048 Boukassa Noma millet 1,000 2,832
NRE080 Yada Iri cowpea 800 2,463 
ggplot(hh[!hhid %in% outlier], aes(costs_ha_ppp, yield_ha_kg)) +
  geom_smooth(size=.8) +
  geom_point(alpha=.7, shape=20, color=1) +
  scale_x_continuous(labels=comma) +
  scale_y_continuous(labels=comma) +
  facet_wrap(~crop, scales="free") +
  labs(x="", y="",
    title="Production Frontier (Output vs. Input) - Niger",
    subtitle="Each point is a respondent. Shade shows 90% CI (kg vs. PPP$ / ha)") +
  theme_def(legend.position="none")

ggplot(hh[!hhid %in% outlier], aes(costs_ha_ppp, yield_ha_kg)) +
  geom_smooth(aes(color=gender, fill=gender), size=.8, level=.9) +
  geom_point(alpha=.7, shape=20) +
  scale_x_continuous(labels=comma) +
  scale_y_continuous(labels=comma) +
  facet_wrap(~crop, scales="free") +
  labs(x="", y="", color="", fill="",
    title="Production Frontier (Output vs. Input) - Niger",
    subtitle="Each point is a respondent. Shade shows 90% CI (kg vs. PPP$ / ha)") +
  theme_def(legend.position="top")

Profitability

Farmers’ gross profit margins by gender and years in seed club.

ttt(margin_ha_ppp ~ group+crop | gender+years, data=hh, render=fmt,
  caption="Mean Gross Profit Margin in Absolute Terms (PPP$ / ha) - Niger")
Mean Gross Profit Margin in Absolute Terms (PPP$ / ha) - Niger
group crop Statistic < 5 ≥ 5
Male Female Male Female
Boukassa Noma millet mean 311 -130 376 54
median 263 -130 163 54
sd 679 44 597 231
Cigaban Matta millet mean 723 216 629 45
median 701 216 386 45
sd 49 NA 796 NA
Yada Iri cowpea mean 130 -16 427 779
median 130 -16 372 701
sd 72 NA 848 188
Himma cowpea mean 80 1,031 1,099 511
median 80 1,031 730 313
sd NA 263 1,088 903
millet mean NA 687 192 398
median NA 687 430 349
sd NA 1,021 433 760 
ttt(margin_ha_sh ~ group+crop | gender+years, data=hh, render=fmt_pct,
  caption="Mean Gross Profit Margin in Relative Terms (% of variable input costs) - Niger")
Mean Gross Profit Margin in Relative Terms (% of variable input costs) - Niger
group crop Statistic < 5 ≥ 5
Male Female Male Female
Boukassa Noma millet mean -7% -53% -12% -41%
median 29% -53% 27% -41%
sd 81% 64% 76% 83%
Cigaban Matta millet mean 333% 86% 176% 30%
median 323% 86% 197% 30%
sd 36% NA 189% NA
Yada Iri cowpea mean 51% -4% 55% 144%
median 51% -4% 46% 151%
sd 50% NA 139% 64%
Himma cowpea mean 11% 175% 140% 52%
median 11% 175% 97% 51%
sd NA 78% 145% 136%
millet mean NA 483% 21% 252%
median NA 483% 30% 190%
sd NA 704% 86% 323%

Note that 24 respondents show negative margins.

kbl(caption="Respondents with negative gross margins.",
  hh[margin_ha_ppp < 0, 
    .(hhid, code, group, crop, costs_ha_ppp, yield_ha_kg, sales_ha_kg, margin_ha_ppp)],
  format.args=list(big.mark=",", digits=1))
Tab. 4: Respondents with negative gross margins.
hhid code group crop costs_ha_ppp yield_ha_kg sales_ha_kg margin_ha_ppp
NRE003 Sabon_Machi_NI_FG1 Himma cowpea 437 700 100 -146
NRE004 Sabon_Machi_NI_FG2 Himma cowpea 280 800 6 -257
NRE007 Sabon_Machi_MI_FG1 Himma millet 229 800 200 -35
NRE008 Sabon_Machi_MI_FG1 Himma millet 487 700 60 -254
NRE009 Sabon_Machi_MI_FG1 Himma millet 490 800 80 -334
NRE010 Sabon_Machi_MI_FG1 Himma millet 551 700 80 -241
NRE011 Sabon_Machi_MI_FG1 Himma millet 326 600 20 -307
NRE016 Sabon_Machi_MI_FG2 Himma millet 1,279 420 250 -793
NRE026 Sabon_Machi_NI_FG1 Himma cowpea 631 800 200 -10
NRE031 Sabon_Machi_NI_FG2 Himma cowpea 571 700 100 -280
NRE038 Sabon_Machi_MI_FG1 Himma millet 736 800 200 -347
NRE039 Sabon_Machi_MI_FG1 Himma millet 269 750 13 -193
NRE042 Chadakori_MI_FG1 Boukassa Noma millet 109 1,500 0 -109
NRE043 Chadakori_MI_FG1 Boukassa Noma millet 102 150 3 -99
NRE044 Chadakori_MI_FG1 Boukassa Noma millet 2,102 700 500 -161
NRE046 Chadakori_MI_FG1 Boukassa Noma millet 343 1,000 0 -343
NRE047 Chadakori_MI_FG1 Boukassa Noma millet 85 100 0 -85
NRE056 Chadakori_MI_FG2 Cigaban Matta millet 291 500 0 -291
NRE063 Chadakori_NI_FG1 Yada Iri cowpea 404 800 100 -16
NRE074 Chadakori_NI_FG2 Yada Iri cowpea 365 800 0 -365
NRE075 Chadakori_NI_FG2 Yada Iri cowpea 644 800 100 -256
NRE076 Chadakori_NI_FG2 Yada Iri cowpea 520 800 0 -520
NRE078 Chadakori_NI_FG2 Yada Iri cowpea 573 800 120 -340
NRE079 Chadakori_NI_FG2 Yada Iri cowpea 367 200 120 -134 
outlier <- c(
  hh[margin_ha_ppp > median(margin_ha_ppp) + 3*sd(margin_ha_ppp), hhid],
  hh[margin_ha_sh > median(margin_ha_sh) + 3*sd(margin_ha_sh), hhid]
)

kbl(caption="Respondents with gross margins > median + 3*sd",
  hh[hhid %in% outlier, 
    .(hhid, code, group, crop, costs_ha_ppp, yield_ha_kg, sales_ha_kg, 
      margin_ha_ppp, margin_ha_sh)],
  format.args=list(big.mark=",", digits=1))
Tab. 4: Respondents with gross margins > median + 3*sd
hhid code group crop costs_ha_ppp yield_ha_kg sales_ha_kg margin_ha_ppp margin_ha_sh
NRE013 Sabon_Machi_MI_FG2 Himma millet 144 900 400 1,409 10
NRE020 Sabon_Machi_MI_FG2 Himma millet 165 600 300 1,582 10
NRE024 Sabon_Machi_NI_FG1 Himma cowpea 720 600 600 2,774 4
NRE025 Sabon_Machi_NI_FG1 Himma cowpea 679 700 600 2,814 4
NRE033 Sabon_Machi_NI_FG2 Himma cowpea 1,007 1,000 1,000 2,875 3 
ggplot(hh, aes(x=hhid, color=group)) +
  geom_hline(aes(yintercept=0), color=1) +
  geom_linerange(aes(ymin=0, ymax=margin_ha_ppp), size=.6) +
  geom_point(aes(y=0), shape=20, size=1.4) +
  geom_point(aes(y=margin_ha_ppp, shape=margin_ha_ppp < 0, fill=group), size=1.4) +
  scale_y_continuous(labels=comma) +
  scale_shape_manual(values=24:25) +
  guides(x="none", shape="none") +
  labs(x=NULL, y=NULL, color="", fill="",
    title="Profit Margin (PPP$ / ha) - Niger",
    subtitle="Each bar is a respondent's gross profit margin") +
  theme_def(
    legend.position="right",
    panel.grid.major.x=element_blank()
  )

Farmers’ gross profit margins by gender and across groups in both absolute terms and in relative terms as percentage of total costs per hectare.

ggBoxTest(hh[!hhid %in% outlier], 
  aes(gender, margin_ha_ppp, color=gender, fill=gender), cp=list(1:2)) +
  scale_y_continuous(labels=comma) +
  facet_wrap(~crop) +
  labs(x="", y="", fill="", color="",
    title="Gross Profit Margin in Absolute Terms - Niger",
    subtitle="Stratified by gender (PPP$ / ha)") +
  theme_def(legend.position="none")

ggBoxTest(hh[!hhid %in% outlier], 
  aes(gender, margin_ha_sh, color=gender, fill=gender), cp=list(1:2)) +
  scale_y_continuous(labels=percent) +
  facet_wrap(~crop) +
  labs(x="", y="", fill="", color="",
    title="Gross Profit Margin in Relative Terms - Niger",
    subtitle="Stratified by gender (% of total costs)") +
  theme_def(legend.position="none")

ggBoxTest(hh[!hhid %in% outlier], 
  aes(years, margin_ha_ppp, color=years, fill=years), cp=list(1:2)) +
  scale_y_continuous(labels=comma) +
  facet_wrap(~crop) +
  labs(x="", y="", fill="",
    title="Gross Profit Margin in Absolute Terms - Niger",
    subtitle="Stratified by years in seed club (PPP$ / ha)") +
  theme_def(legend.position="none")

ggBoxTest(hh[!hhid %in% outlier], 
  aes(years, margin_ha_sh, color=years, fill=years), cp=list(1:2)) +
  scale_y_continuous(labels=percent) +
  facet_wrap(~crop) +
  labs(x="", y="", fill="",
    title="Gross Profit Margin in Relative Terms - Niger",
    subtitle="Stratified by years in seed club (% of total costs)") +
  theme_def(legend.position="none")

ggBoxTest(hh[!hhid %in% outlier], 
  aes(group, margin_ha_ppp, color=group, fill=group)) +
  scale_x_discrete(labels=label_wrap(5)) +
  scale_y_continuous(labels=comma) + 
  facet_wrap(~crop) +
  labs(x="", y="",
    title="Gross Profit Margin in Absolute Terms - Niger",
    subtitle="Stratified by seed club (PPP$ / ha)") +
  theme_def(legend.position="none")

ggBoxTest(hh[!hhid %in% outlier], 
  aes(group, margin_ha_sh, color=group, fill=group)) +
  scale_x_discrete(labels=label_wrap(5)) +  
  scale_y_continuous(labels=percent) +  
  facet_wrap(~crop) +
  labs(x="", y="",
    title="Gross Profit Margin in Relative Terms - Niger",
    subtitle="Stratified by seed club (% of total costs)") +
  theme_def(legend.position="none")

ggplot(hh[!hhid %in% outlier], aes(member_years, margin_ha_ppp)) +
  geom_smooth(size=.8) +
  geom_point(alpha=.7, shape=20) +
  scale_x_continuous(limits=c(0, 22)) +
  scale_y_continuous(labels=comma) +
  labs(x="", y="", color="",
    title="Gross Profit Margin in Absolute Terms vs. Years in Seed Club - Niger",
    subtitle="Each point is a respondent (years vs. PPP$)") +
  theme_def(legend.position="top")

Correlation

Significant pairwise associations.

ggpairs(
  hh[, .(`seed club`=group, `years in club`=member_years,
    `costs PPP$`=costs_ha_ppp, `seed yield kg/ha`=yield_ha_kg,
    `margin PPP$`=margin_ha_ppp, `margin %`=margin_ha_sh)],
  upper = list(
    continuous=wrap("cor", size=4), 
    combo=wrap("summarise_by", color=pal[1:4], size=2)),
  lower = list(
    continuous=wrap("smooth", shape=NA), 
    combo=wrap("box_no_facet", fill=pal[1:4], alpha=.8)),
  diag = list(
    continuous=wrap("densityDiag", fill=NA),
    discrete=wrap("barDiag", fill=pal[1:4], alpha=.8)),
  title="Correlogram stratified by seed club - Niger"
) + 
  theme_def(
    strip.text=element_text(hjust=.5),
    axis.text.x=element_text(angle=-45),
    panel.grid.major=element_blank()
  )

ggpairs(
  hh[, .(gender, `years in club`=member_years,
    `costs PPP$`=costs_ha_ppp, `seed yield kg/ha`=yield_ha_kg, 
    `margin PPP$`=margin_ha_ppp, `margin %`=margin_ha_sh)],
  upper = list(
    continuous=wrap("cor", size=4), 
    combo=wrap("summarise_by", color=pal[1:2], size=2)),
  lower = list(
    continuous=wrap("smooth", shape=NA, color=hh[, pal[gender]]), 
    combo=wrap("box_no_facet", fill=pal[1:2], alpha=.8)),
  diag = list(
    continuous=wrap("densityDiag", fill=NA),
    discrete=wrap("barDiag", fill=pal[1:2], alpha=.8)),
  title="Correlogram stratified by gender - Niger"
) +   
  theme_def(
    strip.text=element_text(hjust=.5),
    panel.grid.major=element_blank()
  )

ggpairs(
  hh[, .(`years in club`=years,
    `costs PPP$`=costs_ha_ppp, `seed yield kg/ha`=yield_ha_kg, 
    `margin PPP$`=margin_ha_ppp, `margin %`=margin_ha_sh)],
  upper = list(
    continuous=wrap("cor", size=4), 
    combo=wrap("summarise_by", color=pal[1:2], size=2)),
  lower = list(
    continuous=wrap("smooth", shape=NA, color=hh[, pal[gender]]), 
    combo=wrap("box_no_facet", fill=pal[1:2], alpha=.8)),
  diag = list(
    continuous=wrap("densityDiag", fill=NA),
    discrete=wrap("barDiag", fill=pal[1:2], alpha=.8)),
  title="Correlogram stratified by years in seed club - Niger"
) +   
  theme_def(
    strip.text=element_text(hjust=.5),
    panel.grid.major=element_blank()
  )

saveRDS(hh, "../tmp/data_nre.rds")