Structure of U.S. Farms

The USDA has a long history of tracking the structure of U.S. farms, and there has been a consistent trend towards a bimodal pattern. Hoppe (Reference Hoppe2014) reported that small family farms make up 90 percent of the U.S. farm count but produce a 26-percent share of farm output, whereas midsize and large-scale family farms account for eight percent of U.S. farms but 60 percent of the value of production. Despite their modest share of farm output, small farms operate half of U.S. farmland and play key roles in several commodities, including producing roughly half of poultry, hay, and other livestock commodities (Hoppe Reference Hoppe2014). Accordingly, there is focused attention on the financial performance of these operations.

Hoppe and MacDonald (Reference Hoppe and MacDonald2015) find a strong relationship between farm scale and operating profit margin (OPM, the ratio of operating profit to gross farm income)—a measure commonly used to assess financial performance. In 2013, 69 percent of all U.S. farms were in the OPM critical zone (defined as less than 10 percent of the farm's gross cash farm income from the sale of commodities and other farm-related goods and services, together with government payments), of these, 87 percent had gross cash farm income less than $100,000. In general, small farms (under $350,000 in gross cash farm income) are less profitable than larger farms due to large farms’ abilities to take advantage of economies of scale. Hoppe and MacDonald (Reference Hoppe and MacDonald2015) provide two key elements of how these small-scale operators are able to remain in business given the large number in the OPM critical zone: (1) undervaluing operator labor and (2) receiving substantial off-farm income. We posit another potential factor: choice of marketing channels.

The USDA ERS, in a 2014 report on farm structure and finances, found that family farms become more diversified as their size increases: between 35 and 58 percent of family farms with gross cash farm income of $150,000 or more produced four or more varieties, compared to less than 10 percent of smaller family farms (Hoppe Reference Hoppe2014). However, they do not consider market channel selection in their consideration of diversification, and market diversification may have a differential influence on financial outcomes depending on the scale of operation, reflecting an important point of consideration. Selling through local markets may be a means through which small-scale operators can secure higher per-unit value for their products (Hardesty and Leff Reference Hardesty and Leff2010, LeRoux et al. Reference LeRoux, Schmit, Roth and Streeter2010).

Exploring Comparative Advantage for Small-Scale Firms

There is a substantial literature that considers comparative advantage for small-scale firms. Hammeresh, Anderson, and Harris (Reference Hammeresh, Anderson and Harris1978) found that a small market share is not necessarily a handicap; it can be a significant advantage that enables a company to compete in ways that are unavailable to its larger rivals. The authors identify and analyze characteristics that help explain how firms with low market share are able to be financially successful and note that these firms compete only in areas where their particular strengths are most highly valued and eschew growth for growth's sake. As most businesses must compete in a limited number of segments within their industry, they must choose these segments carefully: most successful companies define market segments in unique and creative ways. For example, in addition to products and customers, a market can be segmented by level of customer service, stage of production, price performance characteristics, credit arrangements with customers, location of plants, characteristics of manufacturing equipment, channels of distribution, and financial policies.

Kim and Lee (Reference Kim and Lee2010) consider how a responsive supply chain that adds value to customer satisfaction, quality improvements, new product introduction, service level improvements, and lead-time reduction could be a comparative advantage for small-scale enterprises. Strategic collaboration can support a responsive supply chain as collaboration between supply chain partners at the strategic level indicates a strong commitment to buyers and encourages a more responsive supply chain. Value chain models of market intermediation, developed in part to enhance small- and mid-scale producers’ ability to capture price premiums in the marketplace for the environmental, economic, and social benefits embedded in the products, suggest these are some of the actions increasingly present in local and regional food systems (Hardesty et al. Reference Hardesty, Feenstra, Visher, Lerman, Thilmany McFadden, Bauman, Gillpatrick and Nurse-Rainbolt2014).

Farm Debt Usage, Survivability, and Scale

One major concern for the USDA is small family farms’ ability to access and appropriately use capital. In 2014, small family farms held 71 percent of farm assets and 80 percent of farm debt (Patrick, Kuhns and Borchers Reference Patrick, Kuhns and Borchers2016). One solvency ratio commonly used by the USDA ERS, the D/A ratio, was 12.26 in 2015, up from 11.75 the year before (USDA ERS 2016). Hoppe reported that small family farms had D/A ratios between 0.1 and 4.5, depending on farm type. This is compared to mid-sized and large/very large family farms with D/A ratios at 8.4, 12.2, and 21.8, respectively (Hoppe Reference Hoppe2014).

Low et al. (Reference Low, Adalja, Beaulieu, Key, Martinez, Melton, Perez, Ralston, Stewart, Shuttles, Vogel and Jablonski2015) conclude that direct marketing was associated with higher survival rates, a measure of particular interest when considering beginning farms and ranches. One reason for this may be due to lean management strategies. Given, as mentioned above, that 85 percent of farms that participated in direct and intermediated markets in 2012 had gross cash farm income under $75,000, local food system participants may purchase less machinery and land than did those with traditional marketing, and instead, lease such assets as a means of managing cash flows. Consequently, local food market channel participants may not need to leverage as much of their wealth to obtain financing. Lower leverage translates to repayment capacity, thereby reducing the risk of small business failure (Fotopoulos and Louri Reference Fotopoulos and Louri2000).

Profits fluctuate widely in commodity markets because of changes in input and output prices and yields, something that direct marketing strategies can mitigate (through risk-sharing models such as community-supported agriculture or price negotiating options commonly allowed in intermediated markets such as food hubs). Moreover, local and regionally focused producers assume additional supply chain functions (e.g., marketing, processing, distribution) (King et al. Reference King, Hand, DiGiacomo, Clancy, Gomez, Hardesty, Lev and McLaughlin2010). Through taking on these additional roles they receive some of their income from these post farmgate activities; farm income risk may be better managed in these markets.

Previous research on fresh produce price determination found that supply is fixed when price is above marginal harvesting costs, and that any surplus returns above the cost of harvesting are divided among buyers and sellers according to their relative bargaining power in the market (Sexton and Zhang Reference Sexton and Zhang1996). In cases where products are slightly more storable, producers have some ability to control supply, and so the relative bargaining position of growers, shippers, and retailers may vary (Richards and Patterson Reference Richards and Patterson2003). However, Richards and Patterson (Reference Richards and Patterson2003) note that consolidation, vertical integration, and the large scale of many food retailers and distributors are perceived as threats for the producer-supplier's bargaining position, particularly for perishable products such as produce, dairy, and eggs. Accordingly, direct markets are commonly referenced as one mechanism for producers to regain control of their bargaining power by integrating more phases of the food supply chain into their own operations.

Financial Performance and Market Channel

Previous studies explored farm success or financial performance in traditional commodity markets, and in general, financial efficiency, leverage, and farm size were found to be negatively related to mean financial performance (e.g., Ford and Shonkwiler Reference Ford and Shonkwiler1994). Beyond scale, the nature of commodities may also suggest that different marketing strategies will affect market success, and thus, financial viability.

A handful of research uses national data to examine the farm and ranch profitability impacts of sales through local markets. Park (Reference Park2015) examined the impact of participation in direct marketing on the distribution of farm sales using the unconditional quantile regression estimator and endeavored to evaluate the distributional impacts, rather than just assess the means of the sample. Results show lower sales commonly associated with direct marketing enterprises are somewhat mitigated by the scale of the operation; scaling up (at least to some degree, even if not at the scale required to compete in commodity markets) may be an essential element in attaining viability. Park and Lohr (Reference Park and Lohr2010) found that organic producers who participate in local direct marketing channels tend to achieve lower earned income, suggesting that relative dependency on local sales may correlate with negative financial outcomes.

As mentioned above, Low et al. (Reference Low, Adalja, Beaulieu, Key, Martinez, Melton, Perez, Ralston, Stewart, Shuttles, Vogel and Jablonski2015) show that all farms, regardless of scale, are more likely to survive if they have local food sales as part of their marketing portfolio. However, similar to the Park (Reference Park2015) and Park and Lohr (Reference Park and Lohr2010) results, these farms grew at a significantly slower rate compared to farms that did not use these channels.Footnote ¹ Vogel, Jablonski, and Schmit (Reference Vogel, Jablonski and Schmit2016) also use ARMS data to show that small-scale farms using local markets (<$350,000 in gross cash farm income) are significantly more likely to have a positive operating expense ratio and that a greater percentage are profitable than those using only commodity markets. However, they find that this trend reverses once farms exceed $350,000 in gross cash farm income.

Park, Mishra, and Wozniak (Reference Park, Mishra and Wozniak2014) developed a multinomial logit model to assess the relationship between direct markets and financial performance. They find that farmers using direct-to-consumer-only outlets report sales that are significantly lower than from other direct marketing strategies, such as using intermediated only or both direct-to-consumer and intermediated. Of note, they also assess the role that management and marketing skills play in the selection of direct-to-consumer markets. They find that producers with a broader portfolio of management skills—such as more ways to control input costs—were more likely to rely on intermediated retail outlets only or to diversify into both direct-to-consumer and intermediated retail marketing outlets. Finally, a set of exploratory market channel assessments conducted across several states show labor requirements are a major determinant of profitability for small-scale fruit and vegetable producers when comparing different market channels (Hardesty and Leff Reference Hardesty and Leff2010, LeRoux et al. Reference LeRoux, Schmit, Roth and Streeter2010, Christensen et al. Reference Christensen, Thilmany, Jablonski, Sullins and Naasz2017). Because this is a notable pivot from the labor-saving strategies used by commodity operations, the role of labor in financial performance is worth exploring further. In summary, the results from this study provide further evidence of the heterogeneous factors that may drive participants in local food markets.

A Conceptual Model: Benchmarking Financial Performance against Market Orientation

Bauman et al. (Reference Bauman, Shideler, Thilmany, Taylor and Angelo2014) developed a generalized typology of marketing channels that represent strategic market choices or decisions from which farmers would choose depending on the best fit for their business (see Figure 1). The set of marketing strategies is divided into four quadrants using sales volume as the horizontal dimension and value-added (operating profit margin) per unit of sales as the vertical dimension. The types of marketing strategies are ordered to represent common evolutionary steps that operations may take if their current marketing choice or portfolio evolves with plans to expand or decrease in scale, perhaps as new marketing opportunities appear or financial challenges arise.

Figure 1.

Typology of Food Marketing Strategies by Sales Volume and Profit Margin

The business models shown in the top two quadrants of Figure 1 are the focus of this article, because they correspond most directly to the direct-to-consumer and intermediated sales models found in local and regional food systems. Importantly, as reported by Low et al. (Reference Low, Adalja, Beaulieu, Key, Martinez, Melton, Perez, Ralston, Stewart, Shuttles, Vogel and Jablonski2015) and Park, Mishra, and Wozniak (Reference Park, Mishra and Wozniak2014), a portfolio of these marketing strategies, rather than just direct sales, may correspond to a positive influence on the financial viability of a farm enterprise.

The literature on potential comparative advantages sought by small-scale enterprises, more thoroughly described above, suggests that a targeted focus on the market segments they can best serve with responsive supply chains may be key (Hammeresh, Anderson and Harris Reference Hammeresh, Anderson and Harris1978, Kim and Lee Reference Kim and Lee2010). However, research on farm operations participating in direct marketing have reported mixed effects in terms of financial performance and survivability (Park and Lohr Reference Park and Lohr2010, Low et al. Reference Low, Adalja, Beaulieu, Key, Martinez, Melton, Perez, Ralston, Stewart, Shuttles, Vogel and Jablonski2015, Park Reference Park2015). Perhaps that is due to the fact that any analysis of an average producer in this sector may mask the subtle differences among segments that are performing well by participating in direct and intermediated markets. Figure 2 represents one of the many tradeoffs that may be faced by producers who manage their own marketing supply chains.

Figure 2.

Tradeoffs between Efficiency and Customer Service

On the vertical axis, there is a consideration of the scale efficiencies that may be realized (or lost) depending on the volume of sales a market can provide to a farm operation. Further, the sizeable share of small producers participating in direct and intermediated markets who have been labeled as “lifestyle farms” (less than $10,000 in gross cash farm income), may bias findings downward. An analysis of performance by segment facilitates discovery of those who can achieve a viable scale in local and regional markets.

On the horizontal axis, the customer service and responsiveness attributes that may allow farm operations to compete in the marketplace can be varied. Whereas minimal marketing effort may be required for participation in commodity markets, a high level of effort is required for producers who want to better control their prices and customer loyalty. Direct-to-consumer markets are held up as a way for consumers and buyers to reconnect with their food source, support their local economy, and secure higher or more consistent quality products (Thilmany McFadden Reference Thilmany McFadden2015, Vilsack Reference Vilsack2016). Even when scaled up to intermediated markets, research on value chains has begun to show examples of how such assurances may be maintained (Hardesty et al. Reference Hardesty, Feenstra, Visher, Lerman, Thilmany McFadden, Bauman, Gillpatrick and Nurse-Rainbolt2014, Angelo, Jablonski, and Thilmany Reference Angelo, Jablonski and Thilmany2016). Through our analysis of USDA ARMS data, we explore differences across the share of production costs invested in labor as one potential proxy for customer service, given pilot studies demonstrating the essential role of labor in managing responsive supply chains in these marketing channels (LeRoux et al. Reference LeRoux, Schmit, Roth and Streeter2010).

In short, at any point in time a farm operation may be considering the tradeoffs of their product marketing portfolio: balancing the efficiency gains of scale against the potentially higher or more stable returns from self-governed customer responsive supply chains. And, as there is likely heterogeneity among producers participating in these channels with respect to their portfolio mix of markets (illustrated in Figure 1, with the economic tradeoffs presented in Figure 2), an analysis of performance by top- and lower-performing farms may be enlightening.

Although competitive analysis is useful in assessing one's position relative to the competition, it usually does not provide insights as to how competitors achieved this position (i.e., through what methods or processes) without further delineation of benchmark data. By benchmarking several key business factors across producer segments and comparing those operations with relatively better and worse financial performance, we explore several issues of interest. For example, how the relative performance of small and mid-sized enterprises varies from larger enterprises, and if a comparative disadvantage in scale can at least be partially mitigated by participation in alternative marketing channels. Because the cyclical nature of agriculture makes it particularly susceptible to financial risk and repayment challenges, we also assess how farms compare to others in terms of debt usage and financial efficiency (e.g., Fox, Bergen, and Dickson 1993, Patrick, Kuhns and Borchers Reference Patrick, Kuhns and Borchers2016).

Quartile Groupings by Return on Assets

To better capture the sample's heterogeneity, we group local and regional food marketers into high- and low-performing quartiles by profitability and conduct statistical tests across quartiles. There are many measures of profitability, but in this case we chose to use a farm's reported ROA, a commonly used metric to evaluate the profitability of farm businesses, to sort farms into quartiles as a way to disentangle what differences may exist among the most and least profitable farms (e.g., Ho et al. Reference Ho, Newman, Dalley, Little and Wales2013, Wolf et al. Reference Wolf, Stephenson, Knoblauch and Novakovic2016). The use of a “standardized” measure like ROA, may allow some expected and interesting cases (such as lean farms with few owned assets that are aggressively pursuing high-end produce and product markets) to emerge more clearly. In contrast, gross measures (e.g., net farm income) or more refined measures (return on equity) may mask some interesting aspects we expect to see in the comparisons. Beyond sharing the financial profitability differences across these groups, we also explore what operator and production characteristics (region, commodity) indicated to financial outcomes.

Additionally, we examine differences in bargaining power by comparing local producers selling through intermediated channels to similarly sized producers not participating in direct or intermediate markets. Following Cai, Stiegert, and Koontz (Reference Cai, Stiegert and Koontz2011), we use net farm income as a proxy for price, to determine differences in bargaining power between marketing channels. Still, we do present a few other key measures (labor as a share of variable expenses and asset turnover ratios) as part of the analysis to assess how management of labor and assets may also influence financial outcomes.

Profitability by Scale of Farm Sales

Farms participating in direct and intermediated markets report a broad range of profitability with higher performers reporting returns of at least 20 percent, a strong result for a low margin industry such as agriculture (Table 3, last row). Still, over half of the sample reported negative returns, but the losses vary significantly by sales size class. Only the top 25 percent of the smallest size operations (Quartile 4) are profitable, perhaps through adopting a very lean management strategy (an issue we will return to later in the discussion). However, once above the $75,000 sales threshold, all other size classes (Quartiles 3 & 4) have at least one half of their operations reporting break-even profits or some level of profitability, and the highest performing groups report impressive ROA levels of 20, 23, 24 and 29 percent, respectively.

The variance across locally marketing enterprises in the low- and high-income quartiles is very interesting. The lowest performers are likely indicative of the lower left quadrant of Figure 2, lacking the scale economies, and with less-developed supply chains, as one might expect for some small family farms in the United States (retirement operations, or those with off-farm occupations) (Hoppe Reference Hoppe2014). Outliers in all size classes warrant additional exploration in terms of what strategies influence marketing efficiency (vertical axis of Figure 2) or success in terms of supply chain responsiveness (horizontal axis of Figure 2). One additional way to group producers is by the marketing channels in which they participate.

Profitability by Marketing Channel

Table 4 shares the results of ROA-ranked quartiles when farm enterprises are segmented by their marketing choices (direct, intermediated, or some marketing to both types of channels). Again, there is significant heterogeneity within these marketing segments, with the exception of the lowest performing quartiles for direct-to-consumer marketers (where Q1 and Q2 are not significantly different). For the majority of quartiles, the producers using different channels were not significantly different. However, among the top performing quartile (4), direct-to-consumer marketers had significantly lower ROA than the top performers using intermediated markets or both types of markets (not significantly different). This result may signal the importance of intermediated markets to achieve marketing efficiencies in these higher volume channels, if and when responsive supply chains can be maintained through value chain partners.

Table 4.

Return on Assets, by Marketing Outlets, by ROA Rank Quartiles

Note: Return on assets was multiplied by 100 for interpretation.

In a test of whether or not quartiles varied by marketing channel (for example, Quartile 1 farms had significantly different ROA when comparing those using direct vs. intermediated markets), only Quartile 4 was significantly different across marketing channels.

When pairwise tests of mean ROA measures were conducted within each marketing channel, all pairs were found to be statistically significantly different, with the exception of Quartiles 1 and 2 among farms that used both direct and intermediated channels.

Profitability by Marketing Channel and by Scale

Because one could imagine that farm scale may influence the choice of marketing channel, the data were further decomposed to explore how scale and marketing channel interact (Table 5). When testing whether the quartiles varied by marketing channel controlling for scale, there were some interesting patterns that emerged in Quartile 4 (the highest performers), particularly between producers participating in direct and intermediated or both direct and intermediated markets in the two higher sales categories (at least $75,000 in sales). This could signal that once above a certain sales threshold, participating in intermediated markets with more potential for wholesale volumes provides the best opportunities to maximize returns.

Table 5.

Return on Assets, by Marketing Outlets and by Gross Farm Income, by ROA Rank Quartiles

Note: In several cases, the sample size for a scale/marketing channel combination was not large enough to report results (designated by NA).

*In a test of whether or not quartiles varied by marketing channel when controlling for scale, most combinations were not significantly different. Only Quartile 4 was significantly different across marketing channels. Those Quartiles (by scale) that were significantly different across channels (when controlling for scale) at the 10 percent significance level are marked with a *.

^In a test of whether or not quartiles varied by scale when controlling for marketing channel, Quartile 4 was significantly different across all scales for each marketing channel option at the 5 percent level, and marked with a ^. The only other two pairwise significant differences were for Quartile 3 in the direct-to-consumer channel and Quartile 2 in the intermediated channel.

When pairwise tests of mean ROA measures were conducted within each scale and marketing channel combination, all pairs were found to be statistically significantly different at the 5 percent level with the exception of Quartiles 3 and 4 in the direct-to-consumer/$1–74,999 sales group and Quartiles 1 and 2 in the both direct/intermediated channels/$1–74,999 sales groups. Given the high standard errors for these combinations, it is not surprising that pairwise tests were not significant.

As another way to look at these interactions, we found quartiles varied by scale when controlling for marketing channel. Quartile 4 was significantly different across all scales for each marketing channel option at the 5 percent level. Although greater scale provided a clear advantage in intermediated markets among these highest performers, small- and mid-scale producers in the direct and both direct and intermediated segments outperformed the highest-grossing farms. Therefore, it seems scale does matter in the choice of appropriate marketing strategies and the portfolio of channels, at least among the most profitable.

Finally, when pairwise tests of mean ROA measures were conducted within each scale and marketing channel combination, all pairs were found to be statistically significantly different at the 5 percent level with the exception of Quartiles 3 and 4 in the direct-to-consumer/$1–74,999 sales group and Quartiles 1 and 2 in the both direct/intermediated channels/$1–74,999 sales groups (although the latter is likely due to the high standard errors among those groups). The overall variability among producers marketing directly and through intermediated markets makes it clear that one should not develop a notion of an average or representative producer in this space. But, instead, there are likely many “right fit” strategy combinations that make sense. We will now consider a few more factors that may influence performance.

Profitability by Commodity

Table 6 presents a decomposition of quartiles for those producers marketing directly or through intermediated markets, by the primary product they market (they self-report their primary product in ARMS). All quartiles were found to be significantly different at the 1 percent level within each of the commodity categories (with the exception of Quartiles 1 and 2 in the fruit and vegetable, livestock, and dairy producers). The variability of profits is also evident when considering products, though field crops had a relatively tighter distribution across groups. Moreover, across producers of different commodities, the highest performing producers (Quartile 4) were significantly different with fruits and vegetable producers reporting the highest returns among those groups.

Table 6.

Return on Assets, by Commodity, by ROA Rank Quartiles

Note: All quartiles were found to be significantly different at the 1 percent level within each of the commodity categories, with the exception of Quartiles 1 and 2 in the fruit and vegetable and livestock dairy producer groups.

*Across the commodity categories, the highest performing producers (Quartile 4) were significantly different at the 5 percent level, all other quartiles were not significantly different.

Profitability by U.S. Region

To determine if there are any spatial aspects that may influence the financial performance of farms, the sample was segmented by regions at a broad level to maintain sufficient sample sizes (see Table 1 for regional sample sizes). Again, Table 7 shows that, according to pairwise t-tests on quartiles within regions, there were significant differences across almost all quartiles. Across regions, all of the top performing quartiles were significantly different, suggesting there may be better market conditions for this business model depending on location. Specifically, the West and Northeast outperform other regions for the best performers, and the Midwest notably lags behind the other regions.

Table 7.

Return on Assets, by USDA Region, by ROA Rank Quartiles

Note: All quartiles were found to be significantly different at the 1 percent level within region, except Quartiles 1 & 2 in NE and South.

*Across the regions, the highest performing producers (Quartile 4) were significantly different at the 10 percent level; all other quartiles were not significantly different.

Profitability by Metro to Rural Location

As a final approach to explore spatial aspects that may influence the financial performance of farms, the sample was segmented by whether the farm was in a rural, metro or metro-adjacent county (using aggregated USDA ERS rural-urban continuum codes). Again, within segments there were significant differences when assessing t-tests on quartile pairings. Across regions, all of the top performing quartiles were significantly different. Those enterprises located in metro counties significantly outperformed those in areas farther from population centers (Table 8).

Table 8.

Return on Assets, Across Metro-Rural Continuum, by ROA Rank Quartiles

Note: All quartiles were found to be significantly different at the 1 percent level within metro, metro-adjacent and rural categories with the except of Quartiles 1 and 2 in the metro group, which were significant at the 10 percent level.

*Across the regions, the highest performing producers (Quartile 4) were significantly different at the 1 percent level; all other quartiles were not significantly different.

Leverage by Scale of Farm

As detailed in the background research, a key indicator of economic performance may relate to the leverage used to finance the business. It should be noted that this is one of the few factors where quartiles within sales classes were all found to be significantly different from one another when tested as a group (although some quartile pairs were similar), therein warranting additional exploration. Table 9 shows that among the smallest income class (under $75,000), there is the least average reported debt. However, it is unclear if this is due to limited access to capital or the choice of the operator to invest equity capital at this stage of development so that the risk of repayment is not a challenge. The least debt was held by the top quartile that was profitable at this scale, consistent with findings from others showing leverage is detrimental to returns (Ford and Shonkwiler Reference Ford and Shonkwiler1994).

Table 9.

Financial Characteristics, by Gross Cash Farm Income, by ROA Rank Quartile

Note: *Labor as a share of all variable costs was not significantly different across quartiles within sales classes, with the exception of Quartiles 1 and 2 in the $1,000–74,999 sales class, which were significantly different at the 10 percent level. Across sales classes, each of the quartile groups are significantly different at the 1 percent level, so for example, the labor expenses for Quartile 4 farms at each sales level are significantly different.

^The asset turnover ratio was evaluated with pairwise tests across quartiles within sales classes; ^ indicates quartiles that were significantly different than the next highest quartile at the 5 percent significance level. Across sales classes, only Quartiles 2 and 3 are significantly different among groups at the 1 percent level.

#The D/A ratio pairwise tests show that those with a # were significantly different than the next highest quartile at the 10 percent level. Across sales classes, all quartiles are significantly different at the 5 percent level with the exception of Quartile 3.

Once we move to the higher-grossing farm operations, the patterns are more bimodal in nature. In general, the best- and worst-performing farmers (Quartile 1 and Quartile 4 as delineated by ROA) both use relatively higher levels of debt (see Table 9). One could imagine a situation where the poorest-performing operations see debt as a solution for cash flow shortfalls, whereas the best-performing operations see debt financing as an opportunity for faster growth. Given the use of debt by even those performing well, it would suggest targeted, government-backed credit access may be an attractive policy solution to support this sector.

Financial Efficiency

Figure 2 suggests that marketing efficiency is a key aspect of financial performance of farm operations, and although the data did not contain enough information to identify marketing efficiency, the gross sales a producer can create from a farm's assets is a good proxy of overall efficiency. Not surprisingly, asset turnover ratio (ATR), which is the farm's ability to create gross sales effectively, shows some interesting and significant patterns across farms that market directly and are of varying scale. As expected, the highest-performing farms generally have the highest ATRs. The higher ATRs among the strong financial performing operations could be the result of several strategies, including lean management (leasing vs. owning land and equipment), high sales through strong customer loyalty and price points, or intensively managed production models (succession planting, full carcass utilization, season extension). Again, it shows there is likely a diverse set of approaches producers in the direct marketing sector use to be competitive.

There are, however, a few exceptions to the rule. In all but the $75–350,000 gross farm income class, there were some pairwise quartiles with no significant differences. For a subset of farms in each sales class, there appears to be a challenge converting assets into saleable goods, which may again be representative of either beginning farmers or the retirement and lifestyle producers identified by Hoppe and MacDonald (Reference Hoppe2015). For the latter, lifestyle producers may seek out a farmstead and lifestyle with aesthetic or amenity value, but with less concern about actively managing their land, animals, or machinery into saleable goods. One interesting exception is that the poorest-performing quartile in several sales classes have relatively high ATRs, showing relatively high technical efficiency in converting assets to sales but they still lag in overall financial performance. Perhaps it is their relatively high indebtedness that explains their high sales conversion not translating to profitability.

Labor Expenses by Scale of Farm

As previous work by Jablonski and Schmit (Reference Jablonski and Schmit2015) suggests, there are significant differences in the use of labor by those farms that market directly. Previous research on responsive supply chains confirms the varying efficiency in managing labor needs of such business strategies (Hardesty and Leff Reference Hardesty and Leff2010, LeRoux et al. Reference LeRoux, Schmit, Roth and Streeter2010). In the results by commodity groups, we noted that more labor-intensive fruit and vegetable producers reported the highest profitability among the strong-performing Quartile 4. This is one signal that labor expenditures, managed in a strategic manner, can pay off in responsive supply chains.

It appears from Table 9 that labor investments are a relatively high share of total costs as operations grow in scale. This may suggest the hours, skill, and expertise needed to manage responsive supply chains is higher as the volume of sales conducted through direct and intermediated markets increases. This is not surprising, given that the functions of the marketing supply chain, which direct marketing farms integrate into their operations (e.g., distribution, brokering), are relatively well-paid enterprises in the food sector.