Let me say first that I have only begun to read the report cited and that the authors have done some excellent research. The issue is really that outlets like NPR offer snappy headlines and a sound bite or two rather than the entire story. However, it is important that food system organizers communicate more data than that to their market community.

I’ll begin with one of the conclusions from the report:
• It is difficult to draw conclusions about the local economic impact of local foods systems because the existing literature has narrow geographic and market scope, making comparing studies complicated. Data necessary to conduct economic impact analyses are costly to obtain, and researchers have yet to agree on a standard way of accounting for the opportunity costs involved when local foods are produced and purchased or on a standard set of economic modeling assumptions. Many questions surrounding the economic impact of local foods remain unanswered and could be addressed by future research (e.g., Are local food systems good for the rural economy? Might the economic benefits of expanding local food systems be unevenly distributed?)
(The authors do mention that case studies are helpful in local food system research because of the chance for context, but warn that makes generalities difficult.)

here are some other facts from the report:

• Farms selling local food through DTC marketing channels were more likely to remain in business over 2007-12 than all farms not using DTC marketing channels, according to census of agriculture data.

•The significance of local food sales totaling an estimated $6.1 billion in 2012.

For organizers (markets, CSAs, farm stands) the takeaway is clear:

1. We need to collect data and work with those researchers that also want to collect it to paint a more nuanced story of the positive impacts of these channels than were able to be included in the report. Those are not limited to: new product testing, constant cycles of introduction for eaters and producers, the opportunity for attempting small (often risky) pilots for increasing access, educational resources for youth, urban/rural connections and more.

2. That data has to be on the multiple impacts of markets, not just on direct sales. Do farmers meet other buyers (intermediate) at the market? Are other outlets dependent on the market for pick up of their goods? Is it a important way for family members to start working for the farm? What about access to shoppers using benefit program dollars-is this an area of new customer sales that DTC farmers have captured almost entirely (and influenced recent national policy?)

3. A dip in the number of new markets opening or DTC sales flattening for a time (if that is indeed the case) may mean something quite different than the implicit assertion that consumers and farmers are choosing other outlets. Factors may include weather issues, or regulatory pressures (see the fee hike suggested by King County in this story as an example) or farmers unable or unwilling to separate sales outlets when reporting data.

4. An example of how market organizers could help researchers is by gathering anecdotal info for future studies to see if DTC farmers choose autonomy and non-economic benefits over higher incomes as was suggested in the report:

The lower total household income suggests that farmers with direct sales may have had less favorable off-farm income opportunities. If true, this could provide them with an incentive to remain in business even if they have less ability or opportunity to expand production.
Higher survival rates and slower growth for those with direct sales might also be explained by different attitudes toward farm versus nonfarm work. Researchers have found evidence that nonpecuniary benefits from self-employment explain why small business owners remain in business despite earning less income (Hamilton, 2000). There is also evidence that the non-pecuniary benefits to farming (e.g., greater autonomy, independence, and lifestyle factors) are substantial (Key and Roberts, 2009). It is possible that farmers who sell directly to consumers derive greater nonpecuniary benefits from their work—perhaps they enjoy interacting with their customers. This would provide a greater incentive for them to remain in business even with lower business expansion possibilities.

Positive impacts

•The economic benefits of farmers’ markets may also extend beyond multiplier effects, which measure short-term impacts. Lev et al. (2003), for example, found that businesses near farmers’ markets reported higher sales on market days. Not only were these additional sales found to directly support the businesses themselves, but they also generated extra tax revenue for the communities in which the markets were located. Brown (2002) found some evidence that farmers’ markets increase property values in the market district.

•Additionally, farmers’ markets can function as business incubators by providing the infrastructure necessary to build skills and gain business experience (Feenstra et al., 2003; Gillespie et al., 2007). Regular interactions can “generate and circulate knowledge that vendors might use to develop new products and creative ways of marketing them” (Hinrichs et al., 2004: 32-33). Feenstra et al. (2003), for example, explored New York, Iowa, and California farmers’ market contributions to the development of vendors’ capacity as entrepreneurs and found that 66 percent of vendors expanded an existing product line, 50 percent added a new product category, and 40 percent made new business contacts. Sales income may be less important than the skills and business experience developed through participation in farmers’ markets (Brown et al., 2007).

Are Farmers Market Sales Peaking? That Might Be Good For Farmers : The Salt : NPR.

the actual USDA report