New Food Safety Policy Adds to Confusion

Resistance to the National Animal Identification Scheme (NAIS) has been strong. This six-year-old USDA policy initiative to reduce the catastrophic effects to the livestock and meat industry from a major animal disease outbreak has received so much negative reaction from so many and now seems stalled. Now another policy initiative (U.S. House bill HR2749) breezed through the House and passed on the final day before the August recess, less than two months after introduction, and now goes to the Senate. HR2749 passed with little debate, reasonably strong bi-partisan support by today’s standards and very strong support from major food-growing Congressional districts. It is worth exploring the apparent anomaly between reaction to NAIS and HR2749, and examining the technological implications of the HR2749 legislation. 

Strong support for HR2749

Even though there are more than 1,000 bills pending in the current U.S. Congress with the words “food safety”, HR2749, sponsored by representatives John Dingell (D-MI), Henry Waxman (D-CA) and three Democratic colleagues seems to have led a charmed life. It was assigned to only one House committee for review, was passed out of committee with a unanimous voice vote, and moved to the House floor to be passed with little debate and no floor-introduced amendments in a lopsided 283-142 vote with eight members not expressing an opinion.

Contrast the broad support for HR2749 with the opposition juggernaut that currently faces implementation of NAIS. There are a number of reasons for the different reactions. The most obvious difference between these two policies is that HR2749 focuses on food products regulated by the Food and Drug Administration (FDA ), which regulates about 80 percent of the U.S. food supply, and the NAIS comes from the USDA ,which regulates meat, poultry, eggs, fish, and other areas (commodities which are not covered under HR2749). 

Support for changes to the way the FDA regulates its food area has been strong, because the FDA-regulated industries have not only frequently been stung by an apparent unceasing string of high-profile recalls, but the industry stakeholders for these FDA-regulated commodities (fruits, vegetables, and manufactured foods) throughout all production segments from the first mile to the retailers seem to recognize that the status quo just doesn’t work anymore. They feel something needs to be done, and, in general, seem to be backing the regulations put forward in HR2749. Industries affected by NAIS, on the other hand, don't have this broad consensus about the risk their industries face, nor a strong desire to change, and they don’t have clear industry leadership on this topic. This willingness to change accompanied by strong industry leadership is the clear differentiator. 

Other differentiators include:

• HR2749 is a Congressional initiative policy change, and NAIS is generated by USDA staffers without detailed, legitimizing Congressional legislation.
• HR2749 puts most of the responsibility for action on the downstream manufacturer and retailer whereas NAIS places most of the action on the first-mile and upstream livestock producer, who has been very vocal in rejecting the NAIS risk message.
• HR2749 includes clear, detailed exemptions for the very smallest first-mile producer who sells directly to the consumer through roadside stands, farmer’s markets, and CSA’s so that these small growers won’t feel they’re being pushed out of business while NAIS is not so clear that these small producers will be affected.
• HR2749 has had broad industry endorsement and strong industry leadership from all stakeholder groups and is in synch with a parallel industry initiative (the Produce Traceability Initiative – PTI) whereas NAIS has few, strong industry champions and no parallel, detailed, industry-driven program. 

Now that HR2749 has gone to the Senate for consideration, the Senate can either use this bill as its starting point, use a parallel Senate bill (S510 sponsored by Senator Durbin, five Republican Senators and five Democratic Senators) as their starting spot, use one of the other numerous food safety bills pending in the Senate as the stalking horse to move forward, or they can do nothing and let the HR2749 bill die in the 111th Congress.  Of these options, using the S510 bill seems to be most likely, as there are relatively few major differences between that bill and HR2749;  the administration definitely has gone on record that they would do something to change the way the FDA regulates food. 

So, if HR2749 and S510 appear to be the basic grist to be considered first by the Senate and then by a joint committee, let’s look at the basic technology implications. The news here is that from a technology point of view, many of the proposed changes are in line with earlier policies but they are written more crisply and clearly.  However, a few of the proposed changes are broad and sweeping departures from previous policy, and there are some serious technology implications. 

Beyond one-up/one-down

All of the proposed legislation to date, including NAIS, only requires that a member of a supply chain know only the immediate, upstream supplier of the ingredients and the immediate downstream customer that buys the product.  When one-up, one-down was first introduced in the BioTerrorism Record Keeping Act of 2002, many companies thought they only had to keep a list of their suppliers and a separate list of their customers. The FDA published industry guidelines that made it clear that these two lists were insufficient, and that the company needed to be able to accurately describe which specific ingredients are included in which specific finished goods. Most companies can’t accurately meet this objective. Regardless of how well the company is able to preserve ingredient identity across their manufacturing floor, HR2749 ups the ante quite a bit by saying that each member of the supply chain must understand the full pedigree of the product before it reaches them (more than just one-up), and the company must understand the pedigree of who owns and handles the product after it leaves their control (more than just one-down). 

Implementing full upstream and downstream visibility is possible, but unless this extended visibility is handled properly, it will destroy many of the existing commercial relationships that exist today. The retailer who can look upstream to identify exactly which farms are growing the peaches they really want has the ability to by-pass intermediate packers, processors, distributors, and brokers thus disintermediating the supply chain, and creating substantial potential resistance to change. 

Technically, it's possible to provide networks that can provide this basic information when needed and still keep confidential the upstream and downstream ownership information so that supply chains aren't disintermediated.  Such networks have been built for other industries, and certain technology providers are in a position today to be able to deliver this functionality. 

Identity must be preserved across transformations

As mentioned above, the Bio-Terrorism Record Keeping Act of 2002 requires a manufacturer to be able to indicate which specific raw materials or ingredients from exactly which supplier delivery are included in a production batch, even when materials are stored in bulk storage and even when there is rework on the finished product. The difference between this earlier law and HR2749 is that HR2749 is, for the first time, very clear about what is required while the earlier law was ambiguous. The technology implication for food companies is that they need to make sure that their system truly associates specific inputs with specific outputs. 

Increased responsibility for knowing good upstream practices

At no point in previous policy prognostication has a member of the food supply chain been held responsible for an upstream action. Under HR2749 individual food companies will be responsible for knowing that product has been produced according to Good Agricultural Practices, which have yet to be defined. The technology implication here is obvious. The network system that needs to provide the e-Pedigree information before and after manufacture needs also to carry information necessary for the company to determine that good agricultural practices were used while still not disintermediating the supply chain.  This task is trickier, but by no means impossible and this level of supplier compliance information being available to the manufacturer will likely improve food safety. 

Restaurants and retail grocery are now included

In previous laws and rules the retail endpoints were exempt from certain record keeping.  That exemption is lifted in HR2749 and the retail grocer and the restaurant operator will now be responsible for knowing not only the e-Pedigree but also that good agricultural practices were used. 

Full electronic records and immediate access

The legislation is also very clear that the only records that really count are electronic ones and these must be immediately available to appropriate government agencies in the event of an investigation. Technically, this immediate access will require some companies who keep paper and pencil, or fax copies as records to change the way they operation. This requirement will also need some special data confidentiality safeguards to ensure government inspectors see only data relevant to their investigation and not some commercially sensitive data.  Again, meeting this goal is quite possible, but it will need to be done with care. 

What makes all of this new policy highly compelling for food supply chain stakeholders is that, for the first time, FDA will have the teeth to compel companies to act. They will have the right of subpoena, the right of mandatory enforcement and the right of mandatory quarantine. Coupled with a whistleblower protection clause in the legislation, FDA will finally have the teeth it has been requesting to do its job. 

The good news for the operators within the produce industry is that compliance with the current industry-driven Produce Traceability Initiative (PTI) will automatically provide compliance with HR2749 and a likely combined bill that would go to President Obama’s desk for signature.  Other industries regulated by FDA would do well to look at the specifics of the PTI, which has clear action items and clear date guidelines as a solid model. 

The Senate bill S510 is generally in agreement with all of these major policy shifts, so it is likely that if S510 is the bill upon which the Senate focuses as a way to address the issue, that any participant in the food supply chain regulated by FDA will need to prepare for some changes to the way they do business. The Senate bill S510  provides a few policy points that are not in HR2749 including:

• DHS to work with food testing laboratories to provide a network to exchange test data among labs
• Case-level identification of all products leaving a manufacturer
• The DHS Secretary to have, within nine months of enactment, provided completed pilots on a traceability system for at least three commodities that have been involved in a high-profile recall. 

The principal irony between the HR2749 legislative journey and the path NAIS has trod is that HR2749 goes much further in terms of government intervention than either NAIS or the preexisting Bio-Terrorism Record-Keeping Act of 2002 has gone in terms of the requirements for the food industry. HR2749, though, appears to have broad support while NAIS doesn't. HR2749 has now gone to the Senate where it has been read twice and assigned to the Committee on Health, Education, Labor and Pensions. By Thanksgiving 2009, we should know which way the political winds will blow and whether the substantial change embodied in HR2749 will become the law of the land.