Five parties responded to the August 10, 2013 request published in the Federal Register for public comment regarding the guideline for the interviews of Approved Insurance Providers (AIPs) and the survey instruments for insurance agents and insured producers. The five parties were a trade association representing the crop insurance companies, three crop insurance companies, and a trade association representing the crop insurance agents. RMA’s response to the comments submitted by the public will address relevant commentary offered by the insurance company trade association and the insurance agent trade association. The comments submitted by the three crop insurance companies simply stated their support for the comments submitted by the insurance company trade association.
Comments Offered by the Insurance Company Trade Association
Comment: Our previous responses emphasized that the cost of delivery of the crop insurance program is a function of the level and quality of services provided to farmers, which depend on competition in the marketplace. These two aspects of the program cannot be divorced from one another. Any attempt to evaluate the cost of delivery must first establish the level of service that RMA intends for companies and agents to provide to farmers. Economic research indicates that the evaluation of costs when information regarding the quality of service is unavailable or inaccurate can bias the estimation of the parameters of the cost function. Unless a benchmark is established for the level of service that should be provided to farmers, any analysis of costs may result in misleading conclusions regarding the cost of supplying insurance. The only explicit recognition we found relating to this issue was in the Federal Register notice itself.
Response: With respect to the comment that cost of delivery estimation will be biased if no consideration is given to the level and quality of service provided to farmers, RMA does not disagree with that statement. RMA recognizes there are likely different levels of service being provided by agents. As a result the survey collects information on the level of effort and time spent on various activities which can be used to understand the differentiation in the degrees of service provision. Thus, RMA does not agree when the commenter infers that costs cannot be analyzed and such analysis could lead to misleading conclusions.
Notwithstanding the comment, it is also the case that the Standard Reinsurance Agreement (SRA) between RMA and the insurance companies (Approved Insurance Providers, AIPs) establishes the standards of competency to which AIPs, insurance agents, and loss adjusters are expected to adhere. Section VIII of Appendix I, for example, addresses the regulatory duties and responsibilities of the AIPs. This includes the training and proficiency requirements for company employees, agents, agency employees, loss adjusters, and contractors that act on behalf of the AIP. Appendix IV of the SRA provides the AIPs with more specific directives regarding the training and proficiency of agents and loss adjusters. For example, Section II of Appendix IV requires the AIPs to establish an annual Training and Performance Evaluation Plan (TPEP). Among the requirements of the TPEP are that it: (a) address the standards acceptable for agent proficiency testing and satisfactory work performance; (b) describe the internal control measures used to assign work and track, monitor and evaluate work performance against the standards; and (c) describe the additional training and monitoring applicable to any individual who fails to meet the standards. Section II also requires the company to develop a sales training curriculum and a loss adjustment training curriculum, and provides the requirements attendant to each of those curriculums. Thus, contrary to the commenter’s concern that no standards exist, RMA has established certain expectations regarding the quality of service provided to producers by AIPs, agents, and loss adjusters. Beyond addressing competency in the crop insurance delivery system, RMA believes marketplace competition is best suited for determining the level of service that is appropriate for individual policyholders. With no ability to compete on either product or price, customer service is the primary basis on which the delivery system, particularly AIPs and agents, are able to compete.
The data collected in the course of the study will be analyzed to determine if there are significant differences in, for example, time spent by agents for various aspects of service, which may well be correlated with the measure of quality. In addition, data collected from the survey of insured farmers will also provide information as to the level of satisfaction the insured farmers have with the level of services provided by agents. Therefore, while the focus of the data collection effort is not on quality of service, information being collected will allow for some consideration of the quality of service that is provided to insured farmers by the federal crop insurance program.
Comment: As designed, the study appears to be incapable of addressing differences in delivery costs across segments of the program or across the agency force. In addition, despite the stated objectives for the study, the study continues to focus on agents’ accounting costs rather than the opportunity costs associated with the delivery of crop insurance.
Response: This study will capture alternative measures of remuneration to crop insurance delivery resources as opposed to the accounting information readily available in the financials the AIPs filed with RMA and State insurance departments. This will entail a detailed review of the AIP financials, interviews with various stakeholders, surveys of insurance agents and insured farmers, use of competing salary data for insurance agents, and possible adjustments to that data to reflect differences in levels of effort required to deliver the federal crop insurance program. Thus, the study will analyze and provide estimates of the underlying economic costs of delivering the federal crop insurance program as opposed to accounting costs.
It appears the commenter is viewing the survey instrument in isolation to make the inference that the study is focused on agents’ accounting costs rather than opportunity costs. However, the commenter participated in the July 10, 2012 meeting with the industry and received a copy of the presentation given by the contractor. That presentation included an extensive discussion of the study objectives, data sources and technical approaches to be used to assess the economic costs of crop insurance program delivery. For example, the study will evaluate the revenue that could have been earned by agents if they had provided other types of insurance services, which is a measure of opportunity (economic) costs. Among the data sources to be used for that purpose will be wage data compiled by the Bureau of Labor Statistics (BLS). Market data on compensation published by BLS by geography will be used to capture the wages that crop insurance agents could earn in other insurance functions. The survey instrument collects information on hours worked as well, which can be used to capture level of effort.
With respect to the comment that the study does not address differences in delivery costs across segments of the program or across the agency force, the survey is capturing the regions, policy types, policy size, and crop types which are the major dimensions of the program. In particular, the insurance agent survey questions are designed to capture the time and efforts incurred by federal crop insurance agents, possible variations in costs across regions and agent types, and the potential effects of various program changes on insurance agents.
Comment: More generally, the study is designed in a way that suggests that there is a market failure of some sort that results in a difference between the amount agents earn as compared to the cost of providing service to farmers. If the purpose for this study is to investigate this issue, the design of the study needs to address the problem at a conceptual level within the theoretical framework of microeconomics and industrial organization of crop insurance markets. If the purpose for this study is to identify opportunities for cost containment, we reiterate our previous comments that cost containment is no longer an issue due to the cap on total A&O compensation imposed by the 2011 SRA. Agent compensation was capped as part of that same agreement.
Response: We do not agree with the commenter’s premise that the study is predicated on the notion that there is inherently some sort of market failure. RMA has the responsibility to ensure that government payments to AIPs for delivery expenses are sufficient to achieve effective program delivery but are not excessive and make efficient use of taxpayer funds. As the federal government is responsible for compensating the delivery system for its A&O costs, it is appropriate for the federal government to seek the best, most objective information possible regarding such costs, so that it can be a responsible steward of taxpayer funds. The study is attempting to collect objective measures of cost of delivery along a variety of dimensions so as to understand the factors driving the delivery costs. The data that is collected and analysis of that data will drive any conclusions that are made.
Comment: In our previous response we indicated that the objectives of any study should address the major issues faced by the crop insurance program. These included “(1) whether the incentive structure is designed in a way that encourages high participation for all states and commodities, (2) whether agents and companies are being adequately and competitively compensated for their efforts, (3) whether the incentive structure ensures that A&O is sufficient to cover all expenses (as compared to companies absorbing a portion of the cost), and (4) whether the current incentive mechanism has unintended consequences including a shift in A&O between states as commodity prices change.” We also recommended that the study “should also attempt to identify inefficiencies and program management decisions that contribute to the cost of delivering the program.” To all appearances, none of these issues have been addressed in the proposed study.
Response: The primary focus of the study was developed in response to a recommendation made by the Government Accountability Office (GAO) in its evaluation of program delivery expenses (GAO-09-445, “Crop Insurance: Opportunities Exist to Reduce the Costs of Administering the Program”) RMA committed to conducting a study focused on the cost of program delivery. The GAO recommendation and RMA response from the report are reproduced below:
“GAO Recommendation 3: Conduct a study of the costs associated with selling and servicing crop insurance policies to establish a standard method for assessing agencies’ reasonable costs in selling and servicing policies.
RMA Response: RMA agrees that a study of the costs associated with selling and servicing crop insurance policies can be useful in evaluating commission expenses paid by insurance companies and in establishing an appropriate rate of reimbursement. RMA will pursue initiating a study contingent on the availability of funds.”
Comment: The primary concern we have with the design of the study is that it appears to have been designed to produce a foregone conclusion. RMA has previously announced its intentions to use the information obtained from the study in future SRA negotiations:
“With this study, RMA is seeking information regarding the reasonable and necessary economic costs for program delivery which can be used to better establish the level of A&O payments in future SRA negotiations…”
(OMB Package Supporting Statement A, page 10, first paragraph, last sentence)
Response: The sentence cited by the commenter is merely a recognition of the point that the information obtained from the study may be used in future SRA negotiations between RMA and the AIPs over payments for program delivery expenses. The study is being undertaken pursuant to a recommendation from the GAO in its report on the costs incurred by the insurance companies for program delivery, GAO-09-445 Crop Insurance: Opportunities Exist to Reduce the Costs of Administering the Program. In that report the GAO recommended that “The agency should also conduct a study of the costs associated with selling and servicing crop insurance policies to establish a standard method for assessing agencies’ reasonable costs in selling and servicing policies.”
RMA agreed with the GAO recommendation because it has the responsibility to ensure that government payments to AIPs for program delivery expenses are sufficient to achieve effective program delivery but are not excessive and wasteful of taxpayer funds. As the federal government is solely responsible for compensating the delivery system for its A&O costs, it is appropriate for the federal government to have the best possible objective information regarding those costs. There is no presumption on the part of either RMA or the contractor regarding what level of A&O payment is inconsistent with those reasonable and necessary economic costs.
Comment: NCIS has carefully examined the AIP interview guide, Farmer survey, and Agent survey in light of this stated objective. As currently designed, the proposed study is incapable of fulfilling this function. To illustrate this point, we note that Question III. 5. d. of the AIP interview document presumes that A&O is intended solely for agent compensation rather than to cover the full amount of AIP expenses. A related concern was raised during the July 10, 2012 KPMG presentation to the industry. Slide 10 of the KPMG presentation included the following statement:
“Insurance company’s operating expenses are expected to be covered by the A&O allowance.”
While the question in the AIP interview guide is untrue on its face, the statement on slide 10 is true in concept but not in practice. The harsh reality is that A&O has been insufficient to cover industry operating expenses over the entire period from 1997 to the present. Exhibit 5.1 of the 2012 Grant Thornton Profitability and Effectiveness Update reports that the shortfall in A&O payments to companies was 4.7% of gross premium in 2011, 6.9% in 2010, and 5.5% in 2009. Unless the study is designed in an unbiased manner that allows the contractor to investigate whether the amount of A&O paid to the industry covers the reasonable and necessary economic costs for program delivery, the validity of any conclusions from the study can and should be called into question.
Response: The purpose of the interview guide is to serve as a “guide” for the interviews of the AIPs. It is not something that will be provided to the interviewees. Rather, the interviewees will receive only a single page high level document so that they know what topics are to be covered in the interview. The detailed content in the interview guide is material that will be used to guide the discussion with the AIPs and provide a structure that will allow us to learn as much as possible about their views. Shown below is the higher level document that will be provided to the AIP interviewees. The detailed information in what was reviewed by the commenter is information that the interviewer would then use to guide the interview. Not all of the detailed questions would necessarily be posed in each interview. Questions posed would be based on the direction of the conversations. The particular comment about 5.d is not appropriate since the purpose of that question is merely to start a conversation with the AIP to get their view about agent compensation. There is no value judgment intended nor is it assumed that the A&O is only intended to cover agent compensation and nothing else. The commenter appears to have read something into this question which was not intended nor is it something that the AIP would receive.
Outline of the Interview:
We have some general questions on the reporting practice the company follows in filing financials to the RMA (especially the financials contained in the Total Expense Template). Specifically, we want to understand where the compensations paid to independent/captive adjusters and independent/captive agents are recorded.
Comment: Before commencing with its study, the contractor needs to demonstrate the validity of its methodology using hypothetical data (combining with the third-party data sources, if need be). A complete description of the workings of their whole methodology should be made available for public comment before conducting the analysis.
Response: RMA solicited proposals for a study of the reasonable and necessary economic costs of crop insurance program delivery. Following an extensive review of the proposals that were submitted, RMA awarded a contract in March of 2012 for a study of program delivery costs. The technical proposal of the successful offeror was incorporated into the contract and is therefore not a matter for public comment. However, as the contractor intends to conduct a number of interviews and surveys, the study will impose an information collection burden on the general public. The Paperwork Reduction Act (PRA) requires every federal agency to obtain approval from the Office of Management and Budget (OMB), in the form of a control number, before using identical questions to collect information from 10 or more persons. Thus, we were required under the PRA to seek public comment and OMB approval solely for the data collection exercise as it relates to the interviews and surveys. Thestudy design, technical approach, etc. are not themselves subject to the PRA.
Nevertheless, the commenter and others were previously given an opportunity to provide input and feedback on the study design and methodology. In particular, the commenter participated in the industry meeting which occurred July 12, 2012 and received a copy of the presentation given by the contractor. In that presentation we provided an extensive discussion of some of the data sources and techniques to be used to assess the economic costs of crop insurance program delivery, as well as sought feedback from participants. In addition, the commenter included a copy of the presentation and referenced it repeatedly in the comments submitted in response to the July 6, 2012 notice in the Federal Register, Docket ID No. FCIC-12-0007 “Notice of Request for Approval of a New Information Collection”. The commenter has thus already been provided an opportunity to discuss and provide feedback on the technical approach to the study.
Comment: We also wish to register our concerns with the Federal Register notice itself. The reference to agents’ essential expenses suggests that the study is intended to identify non-essential expenses. If RMA is intending to substitute its own judgment for the judgment of agents and farmers, the study should be designed to collect sufficient data to allow RMA to determine which activities are essential and which are not. This is not feasible given the current design of the study.
Response: The purpose of the study is to identify and measure, on a national and regional basis, the reasonable and necessary economic costs required for delivery of the Federal crop insurance program. Broadly defined the three cost centers are compensation to agents, loss adjustment costs, and company overhead. For the costs associated with agent compensation, the study will capture alternative measures of remuneration, for example, competing salary data for insurance agents and adjustments to that data to reflect differences in levels of effort required to deliver the federal crop insurance program. A “line-item” review of expenses to judge which are/are not essential is not part of the study.
Comment: The Federal Register notice also refers to the number of agent visits required for farmers to make informed decisions. Agents need to meet with farmers for a variety of reasons other than to educate farmers about the available options and to sell policies. They meet with farmers to collect essential acreage and production information on policies written through that agent. They also meet with farmers in order to maintain or grow their customer base. The farmer benefits from this activity by becoming better informed on the choices available and by finding an agent who can more effectively provide services the farmer needs. The design of the cost of delivery appears to be predicated on the theory that agents are merely order takers whose costs can be measured in terms of the number of producer visits or the number of policies sold. However, agents also incur expense in serving prospective customers even when the agent is unsuccessful at closing the sale. This is an essential aspect of the business, as Congress recognized when it brought private sector delivery into the program. If these legitimate expenses are excluded from consideration due to improper study design, the conclusions drawn from the study will misstate the true cost of private sector delivery.
Response: There is no intention to eliminate such producer visits from the study. The survey is asking the agents to provide time spent on all aspects of selling and servicing crop insurance policies as well as the percentage of their time spent on all of the activities they engage in.
Comment: The ability to obtain useful information from the proposed surveys and interviews depends on the extent of participation the contractor achieves among the various companies, agents, and producers. The survey and interview instruments should indicate that participation is optional and explain how participation benefits the respondent. If participation is not optional, RMA should specify the legal authority that stands behind the administration of the surveys and interviews.
Response: Participation is not required but there is no reason to affirmatively state that participation is not required. The expectation is that some will want to participate and others will not.
Comment: A critical issue with regard to the information collection process is the confidentiality of private information obtained in the course of the study. The willingness of prospective respondents to participate in the process may depend on whether their information will be available solely to the contractor or shared with RMA. Any private information shared with RMA should eliminate all characteristics that could potentially identify the source, and should preferably be limited to industry wide aggregations. If RMA is to have access to detailed information collected by the contractor, this decision should be clarified in advance.
Response: Neither the contractor nor RMA has any intention to use the information to identify any individual. The contractor will use the data to draw conclusions but will not share any of the individual response data with RMA to the extent it is identifiable to individuals.
Comment: As a final point, RMA should clarify whether the interviews with AIPs will be in the form of written responses or in-person interviews, and whether the AIP has any choice in this matter.
Response: Written responses are not required. The contractor will do some of the interviews in person and some by phone and take notes.
Comments on the Survey Instruments
Comment: Prior to discussing the survey instruments in detail, we note that the documents currently refer to policies for crops harvested in 2011. Each of the documents will need to be updated prior to their release to refer to crops harvested in either 2012 or 2013, whichever is more relevant.
Response: While it may be preferable to update the survey to collect 2012 data, it is not contemplated to do so because the sampling approach has already been executed using information from the agent and farmer populations as of 2011. It would require a significant effort to redo the sample selection and the AIP data collection to the year 2012. The use of 2011 data will not diminish the ability to analyze the costs of the federal crop insurance delivery system.
Comment (brief summary): Relative to the AIP interview guide the commenter provided a number of comments regarding the design of various questions and suggested the guide needed a thorough technical editing.
Response: The purpose of the interview guide is merely to serve as a “guide” for the interviews of the AIPs. It is not something that will be provided to the interviewees. Rather, the interviewees will receive only a single page high level document so that they know what topics are to be covered in the interview. The detailed content in the interview guide is material that will be used to guide the discussion with the AIPs and provide a structure that will allow us to learn as much as possible about their views. Many questions in the interviews are deliberately open ended so as to solicit the point of view and information from the AIPs. Not all of the detailed questions would necessarily be posed in each interview. Questions posed would be based on the direction of the conversations.
Comment: One concern with the farmer survey is how it will be used. It is unclear how the questionnaire responses will be used as a check on agent responses.
Response: A survey will be conducted of insured farmers to whom the sampled insurance agents sell crop insurance to determine the level of service (e.g. number of insurance agent visits, educational services, and other services) that is necessary for the farmers to make an informed decision. Several types of data will be collected. These will include general background of the producers and of their insurance agents. It will also include information about the interaction between producers and insurance agents. Data gathered from the survey of insured producers will serve as a consistency check to information gathered from the survey of insurance agents. To illustrate, do the agents and producers report a similar number of interactions? Are the estimates of time spent discussing insurance options consistent, etc.?
Comment: Another concern is questions that are unclear or omit potentially important information. For example, questions 2.3 and 2.4 ask “What percent of the Federal crop insurance policies you purchased…” It appears that the number from which the farmer is to take a percentage is the number of separate policies the farmer purchased. If the farmer has two policies, one for 10 acres of specialty crops in a dollar plan and one for 2000 acres of corn in Revenue Protection (RP), the response would show the farmer to be 50% specialty crops and 50% other plans, which gives a misleading indicator of the importance of corn and RP to the farm.
Response: The question is asking for the percentages of crops or insurance plans associated with all policies combined. The goal is to understand the percentages of crop insurance policies that a producer purchases for different crops or different types of insurance plans among the total number of crop insurance policies acquired. The responses from the questions are not to be interpreted to indicate any sort of relative importance of a particular crop (e.g., corn) or a particular type of federal crop insurance plan (e.g., Revenue Protection) to the producer.
Comment: Question 2.2 Delete the term “For about”.
Response: We will make the suggested change.
Comment: Questions 2.4 under “Revenue Protection” should mention RP and RP-HPE. “Other plans” should mention AGR and AGR-Lite, which are time consuming plans in some regions.
Response: We will make the suggested change.
Comment: Question 2.5 refers to “Farmer Premium” but should use “Farmer-Paid Premium” instead.
Response: We will make the suggested change.
Comment: Question 3.1 asks “From how many insurance agents did you purchase Federal crop insurance policies for crops harvested in 2011? agent(s).” The question is unclear in that a farmer may have talked with two agents from the same agency but only one signed the contract.
Response: We do not believe the question is likely to cause any issues or confusion as the goal is to understand the typical purchasing relationship between the producer and agent. Even though a farmer may have engaged in discussions with more than one agent, the question asks for the number of agents from whom the farmer purchased the Federal crop insurance policies, not the number of agents the farmer talked with.
Comment: Question 4.2 asks the nature of farmer’s interaction with agent(s) “in percent of total time spent on average across policies.” The preceding quoted expression is unclear.
Response: We will revise the quoted expression.
Comment: Question 4.4 b. asks to quantify the frequency and time spent “Consulting with your agent(s) on other farm programs/issues.” It is important for the contractor to recognize that farm programs and crop insurance are interconnected. The 2008 Farm Bill included programs like ACRE and SURE that covered revenue losses and, in the case of SURE, were directly tied to the MPCI coverage decision of the farmer. There is a strong need (and likely to grow under the next farm bill) for the agent to have expertise in farm programs and to be able to advise the producer how crop insurance and farm programs may complement or substitute for one another. Advising on farm programs is going to be an essential part of the sales effort for many agents.
Response: Agreed, that is why the question is included.
Comment: The main concern with the questionnaire is that it is not clear how the questions are going to address the issue of accounting cost versus economic cost (in other words, opportunity cost). If the contractor believes that other, third-party data sources are going to be used to complement the survey information, the method and data to be used to estimate economic costs should be identified and made available for public comment before conducting the analysis.
Response: The commenter and others were previously given an opportunity to provide input and feedback on the study design and methodology. In particular, the commenter participated in the first industry meeting which occurred July 12, 2012 and received a copy of the presentation given by the contractor. In that presentation we provided an extensive discussion of some of the data sources and techniques to be used to assess the economic costs of crop insurance program delivery, as well as sought feedback from participants. In particular, we discussed the use of competing salary data for insurance agents, and adjustments to that data to reflect differences in levels of effort required to deliver the federal crop insurance program, which will be the basis for assessing the economic costs. The survey instrument will provide information needed to reflect differences in the level of effort, but is not intended or designed to provide the base data assessing economic costs.
In addition, the commenter included a copy of the presentation and referenced it repeatedly in the comments submitted in response to the July 6, 2012 notice in the Federal Register, Docket ID No. FCIC-12-0007 “Notice of Request for Approval of a New Information Collection”. The commenter has thus already been provided an opportunity to discuss and provide feedback on the technical approach to the study. Also, because the contractor will conduct a number of interviews and surveys, we were required under the PRA to seek public comment and OMB approval for the data collection exercise as it relates to the interviews and surveys. The study design, technical approach, etc. are not themselves covered by the PRA.
Comment: In Question 2.1, the term “premiums” should be defined.
Response: We will revise the question.
Comment: In Question 3.3, the term “the average work week” needs to be defined.
Response: We will instead use the phrase “typical work week” but do not believe there is otherwise a need to define the term as it is a straightforward concept.
Comment: Is the time for performing the tasks listed in 3.9 limited to the time expended by the agent directly or does it include the time spent by support staff?
Response: The question refers to “your” time. So that limits the question to the time expended directly by the agent.
Comment: Item b under question 3.9 identifies the percent of time spent on “consulting with producers on other farm programs/issues.” As noted above, farm programs and crop insurance are interconnected, and advising on farm programs is an essential part of the sales effort for many agents.
Response: Yes, and that is why the question is included.
Comment: Question 3.9 should clarify what is meant by “internal” agency compliance. Define “internal” and “compliance.” Indicate whether “agency” refers to RMA or the insurance agency itself. In addition, the question should indicate whether these compliance activities refer to those established by RMA, or are other activities required by the AIP or the insurance agency itself.
Response: The question refers to the activities of the insurance agent/agency. The question will be revised to make this clear.
Comment: The allocation of an agent’s time to the various tasks listed in question 3.9 will depend on the types of crops, insurance plans, and geographical dispersion of the agent’s book of business. If the goal of this study is to gain insight into the cost drivers of the program, the design of the survey needs to be improved to quantify the relationship between costs and the factors driving those costs.
Response: Yes, and the survey is designed to elicit information regarding the factors driving costs such as types of crops, insurance plans, and geographical dispersion of the agent’s book of business. See, for example, questions 2.1, 3.5, 3.11, and 3.13. Analysis of the data provided in response to these questions and others will enable us to quantify the relationship between costs and various dimensions of the program.
Comment: Question 3.10 compares a typical new Federal crop policy with “a property/casualty policy.” Consideration needs to be given to the definition of “a property/casualty policy.” The question should indicate whether this is referring to a new or already existing property/casualty policy. In addition, it should clarify whether the property/casualty policy is comparable to Federal crop policy in terms of premium or other factors.
Response: The question is intended to compare sales of a new crop insurance policy to a new property/casualty policy. This will be clarified by adding the word new in “property and casualty.”
Comment: In questions 3.13 and 3.14, “Revenue Protection” should mention RP and RP-HPE. “Other plans” should mention AGR and AGR-Lite, which are time consuming plans in some regions.
Response: We will make the suggested change.
Comment: In Question 3.14, the terms “little effort” and “a lot of effort” must be defined based on objective criteria such as the amount of time required to sell and service each type. Otherwise, this question is highly subjective and prone to judgmental errors.
Response: For analytical purposes the important question is the relative ranking of the effort required to sell and service the various types of insurance, which is inherently a ranking unique to each individual respondent. We will calculate a mean and standard deviation around the responses and will see whether a significant variation in responses exists.
Comment: Question 3.15 includes this note: “(Note: For the purposes of this question, a new policy is defined as a policy where the agent must build a new Actual Production History (APH), i.e., these new policies require more effort than a typical renewal policy).” The statement that a new policy involves more work should be evaluated as a testable hypothesis rather than a conclusion of the interviewer.
Response: The premise that a new policy will, in general, require more effort than a renewal policy is fairly intuitive. Further, agents consulted in the design of the survey indicated this was typically the case. Thus, we believe it is an entirely reasonable assumption to make.
Comment: Question 3.15 asks for the number of new policies. The phrasing of this question provides no context for evaluation. The response cannot be compared to the response to Question 3.1, which asks for the number of producers rather than the number of policies.
Response: RMA has data on the total number of federal crop insurance policies each agent sells and therefore does not need to ask for the total number. Instead, the question is asking for the number of new federal crop insurance policies the agent sells out of the total number of policies. The response to Question 3.15 is not meant to be directly compared to the response to Question 3.1 that asks for the total number of producers the agent sells to.
Comment: Question 3.16 designates thresholds that distinguish small, medium and large policies. The contractor should explain the choice of the selected thresholds to enable the industry to evaluate their reasonableness. For example, if these are based on the percentiles of premium distribution in crop insurance policies, it would be useful to know which percentiles have been selected. In addition, the question should define the term “gross premium.”
Response: The thresholds were determined based on premium data available to RMA. We do not think it is necessary to provide this level of detail on how the thresholds were derived for the respondent to provide a meaningful response. As regards defining the term “gross premium” it is our expectation that crop insurance agents are already familiar with this concept.
Comment: The design of Question 4.1 raises questions on how the responses to this question will be used. For example, can the contractor demonstrate use of it with some hypothetical data (combining with the third-party data sources, if need be)? We request the opportunity to review their entire methodology in advance of the information collection process to gain assurance that the methodology can be used to extract meaningful information. As this appears to be a critical component of the study, it needs to be able to quantify the relationship between costs and the factors driving those costs.
Response: The commenter and others were previously given an opportunity to provide input and feedback on the study design and methodology. In particular, the commenter participated in the industry meeting which occurred July 12, 2012 and received a copy of the presentation given by the contractor. In that presentation we provided an extensive discussion of some of the data sources and techniques to be used to assess the economic costs of crop insurance program delivery, as well as sought feedback from participants. In addition, the commenter included a copy of the presentation and referenced it repeatedly in the comments submitted in response to the July 6, 2012 notice in the Federal Register, Docket ID No. FCIC-12-0007 “Notice of Request for Approval of a New Information Collection”. The commenter has thus already been provided an opportunity to discuss and provide feedback on the technical approach to the study.
More generally, the technical approach to be used in the conduct of the study is specified in the contract between RMA and the contractor and is not a matter for public comment. As part of the study the contractor will conduct a number of interviews and surveys which will impose an information collection burden on the general public. Thus, we were required under the PRA to seek public comment and OMB approval solely for the data collection exercise as it relates to the interviews and surveys. The study design, technical approach, etc. are not themselves covered by the PRA.
The full cost estimation process will be discussed in the report. At a high level, the expense data RMA collects from question 4.1 is to be combined with compensation data from BLS to develop the economic cost estimate for the agents. In addition, cost information from the AIPs, from their annual financial reporting to RMA, will be included.
Comment: In Question 5.1, the term “policyholder” must be clarified. The response to this question could differ for a large policyholder as compared to a policyholder with median premium in the agent’s book of business. It could depend on the number of policies purchased by each policyholder, the number of acres, the number of crops, and so on.
Response: The number of interactions may vary along any number of program dimensions. The agent is being asked to provide the number of interactions for the typical policyholder. To clarify, the question will be reworded as follows: “On average, approximately how many times do you interact with a typical policyholder in one year?”
Comment: In Question 5.2, how is the term “an average policyholder” defined?
Response: While there is no precise definition for average, we think respondents will understand that it is the average of their own set of clients.
Comment: In Question 6.1 and 6.2, the term “Areas” (see top rows in each panel) is ambiguous. Moreover, it is also not clear how respondents will separately indicate added or saved hours per month, or whether the agent should check at most one box on each row.
Response: It is assumed that respondents will only check one box in a row since it is assumed that the initiative either adds to or reduces hours. Since only one box is checked, then the assumption is that whatever hours are put in the “hours added/saved” column will apply to either the increase or reduction (whichever is checked).
Comments Offered by the Insurance Agent Trade Association
Comment: Supporting our contention is the fact that the surveys and interviews proposed under the latest Federal Register notice are little changed from the last time RMA sought public comment in this regard. It is as if the opportunity for public comment is merely a check-the-box exercise.
Response: Public comment regarding the survey instruments and interview guide was first sought with publication of the Federal Register notice dated January 29, 2013. Two parties submitted comments in response to the January 29, 2013 notice. Remarks submitted by the commenter stated the following: “These comments were initially supplied in response to the related Notice of Request for Approval of a New Information Collection in Vol. 77, No. 130 of the Federal Register, as well as to RMA Informational Memorandum IS-11-003.” The commenter’s remarks later stated: “As such, we would urge that interviews of agency owners be conducted, rather than interviews and surveys of all agents, approved insurance providers, and a sampling of producers. However, if RMA intends to move forward with interviews and surveys, including with approved insurance providers and farmer customers, we would urge that interview and survey questions be published in the Federal Register with opportunity to comment to ensure that the questions posed are as probative as possible in determining an agency’s reasonable costs.” Thus, it appears the commenter did not realize the survey instruments and interview guide had been made available to the public for comment.
Comments provided by the other respondent offered no direct commentary regarding the surveys and interviews. Rather, those comments largely focused on the distinction between economic and accounting costs, along with the level of service provided to policyholders. Regarding economic versus accounting costs, our response to the insurance company trade association under the heading “Design of the Study” addresses this issue. Similarly, we address the level of service to policyholders in our response to the insurance company trade association under the heading “Quality of Service.”
Comment: It is also revealing that, although Federal Crop Insurance delivery centers on quality of service to farmers, the cost of delivery study establishes no standard of service that agents and companies are intended to meet, an absolute prerequisite in measuring proper costs it would seem to us. This current exercise is tantamount to establishing the proper costs of building a home without any specifications.
Response: We have addressed the level of service to policyholders in our response to the insurance company trade association under the heading “Quality of Service.”
Comment: There is also no indication in this cost of delivery study that RMA is focused on Congress's clear objective to encourage greater participation and higher coverage levels; no indication that RMA is interested in maintaining a competitive compensation system focused on quality of service that has been absolutely central to meeting Congress's goal; and no indication that RMA appreciates that A&O costs already exceed amounts reimbursed by RMA notwithstanding the fact that the law requires RMA to fully cover these costs.
Response: The primary focus of the study was developed in response to a recommendation made by the GAO in its evaluation of program delivery expenses (GAO-09-445, “Crop Insurance: Opportunities Exist to Reduce the Costs of Administering the Program”) RMA committed to conducting a study focused on the cost of program delivery. The GAO recommendation and RMA response from the report are reproduced below:
“GAO Recommendation 3: Conduct a study of the costs associated with selling and servicing crop insurance policies to establish a standard method for assessing agencies’ reasonable costs in selling and servicing policies.
RMA Response: RMA agrees that a study of the costs associated with selling and servicing crop insurance policies can be useful in evaluating commission expenses paid by insurance companies and in establishing an appropriate rate of reimbursement. RMA will pursue initiating a study contingent on the availability of funds.”
Comment: RMA's cost of delivery study, at least based on the proposed surveys and interviews, appears to equate what crop insurance agents do every day to a Progressive ad: farmers come into the store, pick a policy in a box off the shelf, Flo scans it, and voila, the farmer walks out fully covered. But crop insurance policies don't come in a box. There are a lot of considerations the agent must ensure the farmer fully takes into account before making a decision. A farmer not only needs to understand his perils and the best crop insurance policies to manage these risks but the producer must also understand how Commodity Title policies under the Farm Bill interact with his or her choice. Someone has to explain all of this and, increasingly, that someone has been the agent, especially as Farm Service Agency (FSA) budget constraints have resulted in office closures and fewer staff. There is also acreage and production information to collect. More often than not, agents must go out to the farmer customer (as well as to potential farmer customers) to meet and discuss these issues. And, agents have to deal with the cost of complying with mounting RMA rules and paperwork which the proposed surveys and interviews do not account for at all.
Response: RMA has the responsibility to ensure that government payments to AIPs for program delivery expenses are sufficient to achieve effective program delivery but are not excessive and wasteful of taxpayer funds. As the federal government is solely responsible for compensating the delivery system for its A&O costs, it is appropriate for the federal government to have the best possible objective information regarding those costs. It is not clear that AIP expenditures reflect the economic costs of program delivery, given the nature of government involvement in the federal crop insurance program. AIPs are unable to compete on either product or price which has potentially dampened the effectiveness of normal market disciplines for assuring cost effective program delivery. To the extent that (a) economic costs differ from accounting costs and (b) AIP expenses exceed the A&O payments, the incentive structure may itself be a key factor. The study will shed light on this issue. The study will also examine factors that may affect the amount of work and level of effort required to sell and service federal crop insurance policies as compared to other lines of insurance.
Comment: In short, RMA appears to be trying to identify how it could do an agent's job more cheaply and more effectively. We question whether imparting wisdom to the private sector about how to do things more cheaply and effectively is the federal government's strong suit but that is the road the SRA has naturally led us down when it first broke from the statutory prescription as well as the longstanding tradition of establishing A&O as a percentage of premium. Now, the government is in the guessing business.
Response: Between reinsurance years 2006 and 2008 the aggregate A&O payments to AIPs more than doubled from $960 million to over $2 billion, driven primarily by higher commodity prices. Much of that additional A&O payment was used to compensate crop insurance agents. Yet it is unclear how, e.g., a doubling of the price of corn leads to a doubling of the economic costs of program delivery, though AIP expenses (accounting costs) reflect such an increase. RMA, the public, oversight bodies, and others were concerned that the large increase in A&O payments provided a windfall to the crop insurance delivery system that was not justified by any significant increase in the underlying (economic) costs of program delivery. RMA has the responsibility to ensure that government payments to AIPs for delivery expenses are sufficient to achieve effective program delivery but are not excessive and make efficient use of taxpayer funds. As the federal government is responsible for compensating the delivery system for its A&O costs, it is appropriate for the federal government to seek the best, most objective information possible regarding such costs, so that it can be a responsible steward of taxpayer funds.
Comment: Equally troubling, there is no indication that privacy will be protected in this process and no clear statement as to whether participation in the surveys and interviews is optional or obligatory and, if the latter, where the statutory authority exists to compel this kind of information.
Response: Participation is not required but there is no reason to affirmatively state that participation is not required. The expectation is that some will want to participate and others will not. Accordingly, in our determination of the sample sizes, RMA explicitly factored into the calculations an expected non-response rate to account for the fact that not all survey recipients will elect to participate in the survey. Neither the contractor nor RMA has any intention to use the information to identify any individual.
[
| File Type | application/vnd.openxmlformats-officedocument.wordprocessingml.document |
| Author | Admin |
| File Modified | 0000-00-00 |
| File Created | 2021-01-30 |