Our audit of the federally funded migrant education program (migrant program) highlighted the following:
Despite recent efforts to improve its oversight of the federally funded migrant education program (migrant program), the California Department of Education (Education) has not provided adequate guidance to the regional offices that administer the migrant program's services. Instead, it has relied largely on the judgment of regional administrators and its individual program staff when making decisions about allowable expenses and financial codes used to categorize these expenses. This lack of formal guidance has created inconsistencies and controversy regarding allowable expenses as well as wide variation in how the migrant program regions classify expenses. As a result, Education's recent calculations of regional administrative costs were flawed and inaccurate. These calculations, as well as recent decisions related to vehicle purchases, have continued to sow discord between Education and the regions. Because of a lack of trust, Education also has had difficulty making productive use of a state parent council whose purpose is to advise and assist the migrant program. Partly because of its past inaction and lack of communication, Education now faces numerous grant conditions and reporting requirements imposed by the federal agency overseeing the migrant program.
The migrant program, which is fully funded by the federal government, provides supplemental education services to migrant children. Children can receive migrant program services if they or their parents or guardians are migrant workers in the agriculture or fishing industries and their families have moved in the last three years for the purpose of finding temporary or seasonal employment. Education receives over $130 million each year to carry out the migrant program. The purpose of the funding is to help migrant children achieve academically despite disruptions caused by repeated moves. Federal law and regulations broadly outline allowable activities and services, depending largely on state educational agencies to define more detailed program guidelines. However, Education has not clearly defined what is necessary and reasonable for a variety of expenditure categories. As a result, expenditures for items such as food, vehicles, and even instruction in music are areas of judgment that can lead to disagreements between Education and the migrant program's regions.
Despite the lack of robust guidance, most of the expenditures we reviewed at eight migrant program regions appear allowable and reasonable. In a review of 320 randomly selected expenditures totaling $12.6 million in migrant program funds, we found six instances for which we question whether the expenditures were allowable or reasonable uses of migrant funds. These six expenditures total roughly $14,800. Half of these expenditures relate to food purchased for a parent conference Education sponsored annually; these food costs totaled $100 per day for each attendee. Also, we observed food costs for a parent meeting in one region that totaled almost $33 per person for breakfast and lunch. The costs were higher than what we would consider reasonable, using the federal per diem rates as our comparison. We questioned the remaining two expenditures because they did not relate directly to migrant students or their identified needs. Further, as part of our review of internal controls and regional applications for funds, we found other questionable expenditures that were not in our sample. For example, we found that in one region a former regional director entered into contracts with janitorial and catering companies that she or her then-husband owned. The payments made to these companies totaled approximately $144,000.
Education presented flawed, unreliable calculations to the federal government regarding the amount of funding spent on administrative costs in its migrant program regions. In response to federal concerns, in January 2011 Education created direct service and administrative cost categories that had not existed before that time. Then, using data from prior fiscal years, Education sorted regional expenditures into these categories retroactively. However, Education did not explain these categories to the regions before 2011. Because Education did not direct the regions to use certain codes for administrative or direct service costs only, some regions charged administrative expenditures to codes that Education later determined were direct service codes. Similarly, some regions charged service-related costs to codes that Education later labeled administrative. Because Education retroactively used codes that did not align with the regions' underlying expenditures, its calculations were unreliable. Even so, the results fed perceptions that regional administrative costs were too high.
Additionally, Education has had internal difficulties that could affect its oversight of the migrant program. Over the past four years, Education's migrant program office has faced a turnover rate that is double the national average for turnover in state and local governments. As a result, staff who have been with the migrant program for a short time have been assigned critical tasks. Further, Education has a fractured relationship with some of its migrant program regions. Regional directors for the migrant program (regional directors) have expressed frustration that Education did not consult them before presenting administrative cost calculations to the federal government. The director for the statewide migrant program agreed that discussions between Education and some regional directors remain unproductive.
Finally, Education has not completed an evaluation of the statewide effectiveness of the migrant program and is hampered from doing so by limited data on program performance. Education has only a draft copy of an evaluation of the statewide migrant program, and the draft report indicates that Education cannot effectively measure about half of the program's target outcomes. The data collected about the migrant program are likely insufficient to thoroughly evaluate the program because only summary-level information about services is collected. Therefore, Education faces challenges in assessing the link between services provided and academic achievement. For example, Education's migrant database records a one-day reading program and a 14-week reading program identically under the same reading services category. Because of its data limitations, Education cannot effectively evaluate the services it provides through statewide contracts or the regional structure used to carry out the migrant program.
To minimize the potential for disagreement over allowable migrant program costs, Education should better define the criteria by which it will consider program costs allowable and include those criteria in the migrant program fiscal handbook it provides to the regions.
To address problems with its methodology for calculating administrative costs, Education should do the following:
To determine if the migrant program is effective, Education should finalize its current evaluation of the program and begin developing the capacity to annually produce a more robust evaluation of the program.
To address a lack of detailed migrant program service and outcome data, Education should either expand the capabilities of its existing statewide databases or implement additional systems that would allow regions to capture more detailed data about migrant students.
Education generally agreed with the report recommendations but took exception to a recommendation that it essentially reverse its previous decision to disallow a vehicle purchase at the San Joaquin County Office of Education (San Joaquin). Because we did not make specific recommendations to seven regions we visited, they did not need to respond in writing to the audit report. However, we made recommendations to one of the regions—San Joaquin—resulting from a particular conflict of interest, and the region agreed that it would implement them.