Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

  • Final Evaluation Brief
  • June 2014

Supporting Handicraft Producers in El Salvador

Despite increased production and modest employment gains, no impact on income

View as PDF

Program Overview

MCC’s $449.6 million El Salvador Compact (2007-2012) funded the $67 million Productive Development Project which included the $55 million Production and Businesses Services (PBS) Activity. The activity, which was implemented in two phases (Phase I: 2009 to mid-2010; Phase II: mid-2010 to 2012), along with the Investment Support and Financial Services Activities, aimed to increase production and employment in the Northern Zone. The activity supported various value chains, with about eight percent of participants being artisans. Support for artisans was designed to help them transition to higher-profit handicraft activities, thereby generating investment, expanding markets, and creating new jobs.

Evaluator Description

MCC commissioned Mathematica Policy Research to conduct an impact evaluation of the handicraft value chain support provided by the PBS Activity. Full report results and learning:

Key Findings

Production and Business Practices

  • The assistance influenced artisans to sell handicrafts and devote more of their own labor to handicraft production.
  • Artisans supported by the activity tried new production techniques and were more likely to report creating environmentally friendly products.
  • As a result of the assistance, artisans sought out new clients in the commercial sector.

Employment and Investment

  • Artisans contracted more paid labor as a result of Phase I assistance.
  • The positive effect on employment during Phase I disappeared by end of Phase II, as control group artisans began employing workers at similar levels as treatment group artisans.

Handicraft and Household Income

  • The evaluation found no impacts of the activity on handicraft income. Structural obstacles to marketing and selling handicrafts may have inhibited positive impacts on artisan’s income.
  • Salaried income decreased for participating artisans, likely due to lost opportunities to earn income outside of handicrafts.
  • Assistance to these artisans influenced them to pursue handicrafts production and sales despite the possibility of earning more income through salaried employment.

Evaluation Questions

The final impact evaluation was designed to understand the impact of the offer of PBS handicraft assistance:

  1. 1
    on production levels, business practice adoption, and product diversification?
  2. 2
    on employment creation and investment?
  3. 3
    on handicraft and household income?

Detailed Findings

Findings from the interim impact evaluation and the final performance evaluation of the handicraft, dairy, and horticulture value chains support can be found under the Production and Business Services Activity evaluation brief.

Production and Business Practices

handicrafts displayed for sale in a store assisted by the activity

Handicrafts for sale in a store assisted by the activity

Artisans in the study are largely married females with a basic education, an average age of 40 and an average of 19 years of handicrafts experience. Production and Business Services assistance most commonly took the form of workshops, seminars, and training sessions. The most common topics discussed during sessions were product design, quality control, marketing, and new technologies. The project also offered donations; however, less than ten percent of treatment group artisans reported receiving donations in Phase I, but 27 percent reported receiving donations in Phase II, including paint, fabric, thread, furniture, and sewing machines.

After PBS assistance, treatment group artisans were 17 percentage points more likely to sell handicrafts and worked over 30 more days per year in handicrafts than control group artisans. Treatment group artisans were 26 percentage points more likely to report trying new handicraft production techniques and nine percentage points more likely to report having looked for new clients in the handicraft sector.

Analyzing trends from 2009 to 2012, it appears that PBS assistance influenced artisans to continue producing and selling handicrafts, whereas artisans who did not receive assistance were more likely to transition out of handicraft production and sales.

Employment and Investment

artisan at work demonstrating to children

Artisan at work

Treatment group artisans invested more in production inputs than control group artisans in Phase I, particularly in paid labor.

Artisans who received assistance generated, on average, 0.14 more full-time equivalent jobs than artisans in the control group during late 2009 and 2010. This is equivalent to over one month of additional full-time labor generated by each artisan. However, this positive effect of PBS on employment decreased in 2011 and disappeared by 2012 as labor contracted by control group artisans increased in 2012 relative to previous years.

Handicraft and Household Income

The evaluation found no impact on net annual productive (handicraft) income, which before the project was less than $50 per month. Obstacles including market access difficulties, limited demand during non-peak months, or the inability of assisted artisan groups to ensure or improve the quality of their goods to secure large orders may have inhibited positive impacts on handicraft income.

In addition, there was a negative impact of PBS assistance on net household income largely due to higher salaried income of the control group relative to the treatment group. These findings suggest that the opportunity cost of handicrafts work—or the salaried income that artisans forewent to pursue handicrafts work—could have outweighed their returns from handicraft sales. Particularly in 2011, there is conclusive evidence that the intervention actually reduced total household income as a result of lost opportunities to earn salaried income outside of handicrafts.

The evaluation points out that even though the treatment group had many years of handicrafts experience, these artisans were artisans with survival-oriented businesses who did not possess the relevant skills to meet current market demand for handicrafts. The Phase II implementer stated that these types of artisans have limited potential to succeed in handicrafts and should be provided with technical assistance in sectors outside of handicrafts.

MCC Learning

  • Use the program logic to ensure that the evaluation has sufficient statistical power to track realistic impacts.

  • Test assumptions around the appropriate content and duration of trainings to maximize impact.

Evaluation Methods

Evaluation methodology

This final handicraft impact evaluation used a randomized methodology with an exposure period of one to two years.

To identify eligible participants, the implementer conducted a census of self-identified artisans in the Northern Zone of El Salvador. Because implementation plans required that groups of producers be provided with services at the same time, the impact evaluation randomized municipalities. A total of 19 municipalities, 9 in treatment and 10 in control, were randomized into the study. Originally, the evaluation was designed as a randomized roll-out and the artisans in the control municipalities were scheduled to receive services in Phase II of the project. However, the implementer changed the eligibility criteria for Phase II, and the control group did not meet the new criteria. Therefore, the control group remained untreated.

A survey designed for this evaluation called the Productive Development Survey was conducted by the Salvadoran Direction General for Statistics and Census. The baseline survey was conducted in October 2009 with an interim follow-up survey in November 2010. Results from the interim survey were published in an interim impact evaluation report in 2012. Two additional follow-up surveys were conducted in September 2011 and August 2012. The sample for the evaluation includes all survey respondents that completed all four in-person interviews. This is a total of 587 individuals – 289 in treatment and 298 in control.

Because less than 200 of the 635 artisans who participated in the project in Phase I – and none of the 468 new participants who join in Phase II – are included in the study sample, the evaluation findings are not generalizable to the entire population of artisans who received some form of assistance, particularly those artisans who received assistance in Phase II under more restrictive selection criteria.