Philip G. Pardey University of Minnesota

Transcription

1 Julian M. Alston University of California, Davis Philip G. Pardey University of Minnesota Presented at the Festschrift conference on New Directions in Fight Against Hunger and Malnutrition held in honor of Per Pinstrup-Andersen Cornell University December 12 14, 213

2 The Issue Problem #1 Very high marginal social returns to agricultural R&D Slowing farm producivity growth and rising food commodity prices è Paradox Evidence calls for a significant increase in funding We observe reduced agricultural research investments by high- income countries è Conundrum Do policymakers not understand the evidence and its implica=ons? Are money metric (economic surplus) measures not effec=vely meaningful? Problem #2 Many of the world s poor remain undernourished High and rising rates of obesity and overweight è Policy Conundrum Implica=ons for agricultural research priori=es? To what extent has agricultural R&D contributed to increasing obesity while reducing poverty? Should priori=es emphasize poor farmers in marginal areas or more nutri=ous food? What about Problem #1? What about opportunity cost?

3 Who saved a billion lives? Norman Borlaug has already saved more lives than any other person who ever lived.... The form of agriculture that Borlaug preaches may have prevented a billion deaths. Gregg Easterbrook, The Atlan+c Monthly, 1997 Norman Borlaug! What was the benefit-cost ratio? Alternatives to money metric Measures?

4 A slowdown in producivity growth: cereals (percent) (percent) Yield Price

5 Gains from agricultural TFP growth in 29 (by income group of countries) Income Group TFP in =1 AgGDP Popula=on GDP per capita Benefits Per capita benefits Benefits as % of GDP (all countries) Benefits as % of GDP (TFP growth) index billion 25 PPP$ million 25 PPP$ billion 25 PPP$ billion 25 PPP$ percent percent World 225 3,658 6,796 9,317 1, High income ,82 32, Upper- middle income 228 1,648 2,575 7, Lower- middle income 175 1,266 2,369 2, Low- income , Source: TFP from Keith Fuglie. Benefits computed as B = AgGDP*(TFP 1)/TFP

6 Gains from agricultural TFP growth in 29 (by region) Region TFP in =1 AgGDP Popula=on GDP per capita Benefits Per capita benefits Benefits as % of GDP (all countries) Benefits as % of GDP (posi=ve TFP) index billion 25 PPP $ million 25 PPP$ billion 25 PPP $ billion 25 PPP$ percent percent World 225 3,658 6,796 9,317 1, High Income ,18 33, EE & FSU , LAC , MENA , Asia & Pacific 211 1,978 3,557 4,171 1, SSA ,

7 What s agricultural producivity worth? Output gain per capita (25 PPP$) 9 4 9,317 Asia&Pacific EE&FSU High Income LAC MENA SSA 266 2, 4, 6, 8, GDP per capita (25 PPP$ per person)

8 Gains from agricultural TFP growth in ,317 Output gain as percent of GDP, log scale Asia&Pacific EE&FSU High Income LAC MENA SSA ,981 22,26 162,755 GDP per capita (25 PPP$ per person), log scale 2.9

9 Direct and Indirect Effects of InnovaIon on Poverty... a naïve targeting of research resources on poor producers based only on direct benefits may produce quite misleading results on overall poverty alleviation effects. Byerlee (2) Food Policy, p. 442 frequency poverty line Direct Effects benefits to poor farmers who adopt Indirect Effects: (a) Induced Output Price Changes benefits to consumers reduced benefits to farmers who adopt cost to farmers who do not adopt net benefits to farmer household depend on whether they adopt and whether the household is a net surplus or deficit producer Indirect Effects: (b) Employment and Wage Effects in agriculture in other sectors (a) (b) Where will the world s poor be found in 23? 25? On subsistence farms? In cities? annual income As the world population is becoming increasingly urbanized, the role of technological change in reducing aggregate poverty correspondingly evolves from direct to indirect effects. De Janvry and Sadoulet (22) J. Dev. Stud., p. 4 Will poverty be reduced more effectively by providing technologies targeted towards poor subsistence farmers? technologies that result in more abundant, cheaper food?

10 Marginal share DistribuIon of an addiional 1$ of income across 144 countries, broad categories Zimbabwe World.315 Low income.485 Food, beverage and tobacco Furniture, housing, and clothing Recreation High income.24 Per capita income Medical and health Transportation and communication Other Qatar

11 Marginal propensity to consume food Marginal share of food, bev. and tobacco Marginal share on medical and health Asia&Pacific EE&FSU High Income LAC MENA SSA

12 Agriculture, food and health anthropometric angles

13 Average life expectancy at birth by world region, Years Asia&Pacific EE&FSU High Income LAC MENA SSA

14 Death caused by nutriion related diseases and by age category, 28 High and upper middle income Low and lower middle income Prematurity and low birth weight Million Hypertensive heart disease Ischaemic heart disease Diabetes Million Nutritional deficiencies Diarrhoeal disease 4 2 Cerebrovascular disease >59 All age >59 All age

15 Number of deaths by disease, income group and age group in 28 Cancer Cardiovascular and diabetes Number of deaths 2, 1,5 1, 5 High income Upper middle Lower middle Low - 14 >6 Number of deaths 6, 4, 2, High income Upper middle Lower middle Low - 14 >6 HIV, TB and Malaria Maternal, perinatal, nutritional and diarrhoeal Number of deaths 1, High income Upper middle Lower middle Low - 14 >6 Number of deaths 1,5 1, 5 High income Upper middle Lower middle Low - 14 >6

16 Agricultural policies, household resource allocaion and nutriion HOUSEHOLD RESOURCES AGRICULTURAL PRICE POLICIES FIXED CAPITAL LABOR (TIME ALLOCATION: ENERGY USE Food imports/exports Storage/buffer stocks Price controls Credit Programs Input subsidies Land Animals Physical Capital Educa=on Experience Own- farm produc=on Off- farm produc=on Leisure Household chores (including child care and food prepara=on EXOGENOUS IN THE SHORT RUN FOOD PRICES AND WAGES Own-farm production and income TOTAL HOUSEHOLD INCOME Off-farm income INVESTMENT AND CONSUMPTION EXPENDITURES F E R T L I T Y Other goods and services Housing Health and sanita=on Food AGRICULTURAL INVESTMENT POLICIES Research and extension Irriga=on Roads and other infrastructure ENDOGENOUS TO HOUSEHOLD DECISIONMAKING PROCESS INTRAHOUSEHOLD FOOD DISTRIBUTION INDIVIDUAL NUTRITIONAL STATUS MORBIDITY COMMUNITY HEALTH AND SANITATION Lagged effect

17 How to spend $1 billion????? Agricultural research maximum payoff Policy ObjecIve Maximum Economic Surplus Incomes of farm & nonfarm poor Incomes of poor farmers NutriIon of the poor Reduced obesity Infrastructure (roads, water supply, markets) EducaIon Public health (clean water, sewerage, immuniza=on deworming) NutriIon subsidies (food subsidies, supplements) Effective? è Positive entry NutriIon research Medical research Agricultural research micronutrients Agricultural research healthy foods Efficient? è Cost effective relative to alternatives Agricultural research poor farmers?????

18 A Trade-off of Equity and Efficiency Equity (V) IC 1 V** V max a d V* b BTC* IC BTC R V min c E min E* E** E max Efficiency (E)

19 Agricultural R&D as Obesity Policy? Parks, Alston and Okrent (213) $166.2 billion in additional health-care expenses attributable to obesity 15.2 % of U.S. public medical expenditures in 29 DWL (c.f. BMI* = 25) of $148.2 billion Marginal excess burden is $27 per year per additional unit of adult BMI $3.6 per year per additional pound of adult body weight Alston, Okrent and Parks (213) use these to compute benefits from hypothetical changes in U.S. public agricultural R&D taking into account effects on Prices of food commodities => U.S. food consumption => Obesity patterns Public health-care costs

20 Projected Changes in Calorie ConsumpIon & Body Weight from Changes in Public Research Spending é 1%, all commodi=es é 1% specialty crops ê 1% for all others Revert to 198 knowledge Stock Change in consump+on (kcal/day) Change in body weight (lb/u.s. adult) One year Steady state Source: Alston, Okrent, and Parks (213).

21 Projected Changes in Social Welfare é 1%, all commodi=es é 1%, specialty crops ê 1%, all others Revert to 198 knowledge Stock Change in Steady- State Body Weight for US Adults (lb per capita) Change in Public Health- Care Costs ( H) (millions of $ per year) 3,849-1,634-23,94 Change in Social Welfare (millions of $ per year) Excluding H 28,739-15,63-223,89 Including H 24,89-13, ,95 Cost per Pound Decrease in Body Weight ($ per lb) Excluding H Including H Source: Alston, Okrent, and Parks (213).

22 ImplicaIons Agricultural R&D is likely to be compara+vely ineffec+ve as an instrument of obesity policy it takes a very long =me to take effect economically inefficient R&D yields a very high social payoff we are already inves=ng much too linle high opportunity cost è high net social cost: over $6/lb reducion in adult weight! If you want to use high prices to reduce obesity, more economic to tax food according to its caloric content (Okrent and Alston, 212): calorie tax ($.163/1, calories): sugar tax ($2.64/kg) : $1.77/lb net benefit $1.73/lb net benefit food tax (4.97% ad valorem): $1.54/lb net benefit fat tax ($5./kg): $1.31/lb net benefit Net benefit (negaive net social cost per pound) from introducing food taxes compared with net cost from reducing R&D to reduce obesity

23 Strange synthesis, second- best soluions First- best soluion R&D as an instrument of science policy to correct for market failures in R&D Second- best soluion modified R&D policy, correcing for market failures in R&D with allowance for implicaions for reducing other distorions Third- best soluion recognizing insituional failure) Given gross underinvestment in it, R&D may be a least- cost (or, at least, preny good) instrument for other objecives (related to the environment, health, income distribuion) What is the relevant counterfactual alternaive? Less R&D? An increase in R&D spent sub- opimally may be bener than a coninuing decline in total resources?

24 Equity and Efficiency: Win-Win! Equity (V) V max a V BTC R V min b E min E E max Efficiency (E)

25 Main Points Agricultural R&D has paid handsome dividends Favorable BCRs Increased availability of food and much lower prices Important engine for poverty reducion and growth Current concerns Underinvestment in agricultural R&D Slowdown in spending, especially in high- income countries Diversion of funds away from farm producivity enhancement Slowdown in agricultural producivity growth in many places

26 Main Points Targeted research? (e.g., poor, resource- poor farmers on marginal lands) Will targeted research be effecive? Will it yield (much) smaller total benefits? Will it yield larger or smaller impacts on <poverty reducion>? Is it the least- cost way of achieving the <income distribuion> goal? Doing well by doing good (Derek Tribe) VS Doing good by doing well Can jusify (and perhaps achieve) an increase in R&D funding based on its impact on social objecives, and anain significant net benefits It will osen be best (for equity as well as efficiency objecives) to spend given R&D resources where the total payoff is highest