Summary
When Minnesota became a state in 1858, the federal government granted it sections 16 and 36 of every township, or their equivalent, for the benefit of schools. <FN- The original federal school land grant consisted of 2.9 million acres of land. The state later added swampland and other land grants to the original grant for a total of 8.1 million acres.> The Minnesota Constitution established the Permanent School Fund (PSF) to ensure a long-term source of funds for public education in the state. The PSF consists of the accumulated revenues generated from the land. The state holds the land and accumulated revenues from the land in trust for the benefit of public schools in Minnesota.
The Department of Natural Resources (DNR) is responsible for managing school trust land, much of which had been sold by the mid-1880s. DNR currently manages about 2.5 million acres of school trust land.
The principal of the Permanent School Fund consists of cash generated from the trust land. <FN- The principal of the PSF does not include the value of the trust land.> Income is primarily earned from land and timber sales, land leases, and mineral royalties. The State Board of Investment (SBI) is responsible for investing the PSF principal, which had a market value of about $437 million on June 30, 1997. Interest and dividend earnings are distributed to school districts each year. During the 1995-96 school year, nearly $31 million (less than 1 percent of all state revenues to K-12 schools) was distributed to schools.
In May 1997, the Legislative Audit Commission directed our office to study the state’s management of school trust land and the Permanent School Fund. We asked the following questions:
How has DNR managed the school trust land given its fiduciary responsibilities related to the Minnesota Constitution and state laws?
How does DNR balance its fiduciary responsibilities to the Permanent School Fund with its natural resource management responsibilities?
What returns have timber sales, mining rents and royalties, lakeshore and other leases, and land sales realized for the principal of the PSF?
Do DNR’s administrative costs reflect the actual costs of managing school trust land?
How does Minnesota’s oversight of school trust land management compare with other states? Could another unit of government manage Minnesota’s school trust land more cost-effectively than DNR?
To answer these questions, we used several different DNR databases to examine the characteristics of Minnesota school trust land, estimate the value of timber on commercial forest trust land, and analyze timber sales. We interviewed staff from the Department of Natural Resources, the State Board of Investment, Minnesota county land departments, and land management and fund investment agencies in other states, and members of the Permanent School Fund Advisory Committee. We reviewed literature, state laws, and case law related to management of school trust land.
The federal government’s grant of land to Minnesota "for the use of schools" and the state’s acceptance of the grant created a trust. <FN- A trust is a right of property held by one party, a trustee, for the benefit of another.> When the State of Minnesota accepted the terms and conditions of the federal land grant, it accepted the position of trustee for public schools in Minnesota. The trustee relationship extends to the Minnesota Legislature, DNR, SBI, and other state officials who make decisions affecting the trust. DNR officials told us that their actions as land management trustees are often constrained by conflicting legislative direction and limited funding.
We recognize that the interests of the trust can coincide with the general interests of the state, and that state actions are often consistent with the interests of both the trust and the general public. Nevertheless, state officials need to be mindful that when their actions affect school trust land, they have special obligations. According to our interpretation of case law, the trust status of the federal school grant land imposes obligations and constraints on how the state may manage school trust land that would not apply if the state held the land outright. The same fiduciary principles that govern the administration of private trusts apply to trustees of school trust land and funds. Case law emphasizes that the trustee’s primary responsibilities are to manage the trust in the interests of current and future beneficiaries. <FN- There have not been any court cases on the nature of the trust relationship in Minnesota. We examined cases from federal district courts and circuit courts of appeals that would be applicable to Minnesota.> The basic long-term objective of the trust should be to generate as much revenue as possible to aid public education.
Minnesota’s 2.5 million acres of school trust land are located primarily in the northern part of the state, as shown in Figure 1. More than 92 percent of school trust land is located in 10 counties: Koochiching, St. Louis, Itasca, Lake, Cass, Aitkin, Cook, Beltrami, Roseau, and Hubbard. School trust land represents 46 percent of the 5.4 million acres of state-owned, DNR-administered land in Minnesota.
Table 1 shows that about 1.9 million acres of trust land are in state forests, state parks, wildlife management areas, and other DNR management units. The Division of Forestry manages 94 percent of the school trust land: 67 percent of the trust land that is in state forests and another 27 percent that is not located in any management unit.
Minnesota has about 2.5 million acres of mineral rights on school trust land and about 1 million acres where the state has "severed mineral rights." Severed mineral rights occur when the state sells the land but retains the subsurface rights. Since 1901, the state has reserved mineral rights when state-owned land is sold.
The PSF principal receives income from economic activities on trust land-mining rents and royalties, land sales, and forest management activities, which include timber sales and leasing of trust land. We found that:
Figure 2 shows the revenues from land management activities added to the PSF principal since 1986. Management of school trust land contributed about $41 million to the PSF between 1986 and 1997, or an average of $3.4 million per year. Mining rents and royalties have provided the most stable source of revenue, generating an average of $1.4 million in revenues each year. Net revenues to the PSF from timber sales and land leases increased from zero in 1991 to nearly $1.7 million in 1997. The lack of net revenues from timber sales prior to 1992 was the planned result of DNR implementing a reforestation policy. Trust land sales, most notably the legislatively-initiated sale of lakeshore lots, represented an increasing revenue stream starting in 1988 and will continue for 20 years as payments are received.
In 1985, the Legislature adopted the following goal for management of school trust land:
It is the goal of the permanent school fund to secure the maximum long-term economic return from school trust lands, consistent with the fiduciary responsibilities imposed by the trust relationship established in the Minnesota Constitution, with sound natural resource conservation and management principles, and with other specific policy provided in state law. <FN- Minn. Stat. §124.079.>
In addition to maximizing the long-term economic return, this goal allows DNR to manage school trust land to serve the public benefit by providing recreational opportunities, wildlife habitat, and other values consistent with natural resource management principles. If trust land is used for purposes that either restrict or prohibit revenue generation, DNR has recognized that it should seek a method of compensating the trust for the foregone revenues.
Our review of DNR policies shows that:
State laws governing the management of forestry, minerals, parks and recreation, and other resources generally apply to all state-owned land, including school trust land. School trust land is managed according to the plans for the management unit in which it is located. In some cases (timber sales and mineral leasing), the plans are consistent with the goal of securing the maximum long-term economic return from trust lands. In other situations (state parks and wildlife management areas), managing for natural resource considerations has the potential to restrict or prohibit economic activities on trust land.
We also found that:
For example, DNR requires that all leases on trust land be charged a cash rental. This means that some contracts allowed on other types of land are excluded from trust land.
All school trust land is not of equal value and does not have the same capacity to generate revenue for the trust. The Division of Forestry maintains a forest inventory database that we used to describe the characteristics and estimate the value of timber on commercial forest trust land. We found that:
About two-thirds of the commercial forest trust land (963,000 acres) has an "excellent/good" timber productivity and one-third (528,000 acres) has "medium/poor" timber productivity. The remaining forest trust land (746,000 acres) is unlikely to yield commercially viable timber harvests because it is: (1) commercial forest land that is not available for harvest because of policy considerations (shoreline setbacks, old growth timber); (2) unproductive because it is inaccessible and swampy; (3) not stocked as forest land; or (4) used for agricultural, industrial, or recreational purposes.
The Minnesota Constitution and state law allow DNR to deduct the costs of managing school trust forest land from the revenues earned by this land. Timber sales from trust land represent the bulk (about 85 percent) of the forest management revenues, but 50 percent of the revenues from lakeshore lot leases and revenues from other leases of school trust land are also used to finance forest management costs.
From 1983 until 1992, forestry management costs for school trust forest land exceeded revenues earned from that land and no income was deposited in the Permanent School Fund. According to DNR, prior to 1983 funding did not permit the department to reforest trust land located outside of state forests. This created a large backlog of forest improvement work (such as reforestation) for subsequent years. In 1992 and thereafter, revenues from managing school trust forest land have exceeded costs (see Figure 3).
Forestry management costs eligible for reimbursement include fire protection, improvement, administration, management, and forest road construction and improvement. <FN Minn. Stat. §16a.25, subd. 5.> We examined how DNR determines its costs for managing trust forestry land and we concluded that:
DNR uses a number of methods to allocate its forest management costs to the trust. The bulk of the costs--managing timber sales and reforestation--are allocated based on the percentage of total timber sale revenues that are generated from trust land. We think using the trust’s proportion of timber sale revenues is a reasonable way to allocate the actual forest management and improvement costs incurred.
DNR provides fire protection services on over 22 million acres of land--about 2.5 million acres of school trust land, 2.9 million acres of other state-owned land, and more than 17 million acres of other public and private land. The costs of fire protection are allocated on a per acre basis. Since school trust forest land represents about 10 percent of the land receiving fire protection, DNR allocates 10 percent of its total fire protection costs to school trust land. However, only about 7 percent of the fires over the past 10 years occurred on all state-owned land. Since school trust land represents about half of state-owned land, it could be assumed that roughly 3.5 percent of all fires (or about half of what occurred on all state-owned land) occurred on trust land. Probably fewer than 3.5 percent of fires occurred on trust land because only 2 percent of the all fires occurred in the counties where trust land is concentrated. Some fire costs are associated with the number of fires, however, DNR believes that the costs of fire prevention and suppression on trust land may be greater than the number of fires would suggest. We recommend that:
DNR currently include the revenues and associated management costs from recreation management (primarily state campgrounds) in the Forest Suspense Account. From 1992 to 1997, the costs charged for recreation management have exceeded the revenues. The Legislature and DNR should consider whether costs for recreation management should continue to be paid from trust revenues.
With DNR’s assistance, we estimated the net income likely to be produced from commercial forest trust land over the next 40 years. Like any model, our estimate of timber value relies on a number of simplified assumptions. For example, we assumed that the state and the trust’s share of the total timber harvest would remain the same over the 40-year period as it is today. We estimate that:
These estimates are sensitive to assumptions about DNR’s forestry management costs. If DNR management costs increase at a rate of 2 percent more than inflation over the next 40 years, then the estimated present value of net income from timber decreases to between $93 million and $213 million. These estimates of the timber harvest for trust land are similar to rough estimates of the value of commercial forest trust land provided by DNR. Valuing trust commercial forest land at $300 million, the return on asset value from timber sales on school trust land has been less than 1 percent in recent years. DNR hopes past investments in reforestation and forest management will increase future returns.
We also examined what portion of timber sale activity is generated from school trust land. Timber sales on trust land accounted for over 50 percent of both the volume and value of state-owned timber sold between 1986 and 1996. Between 1955 and 1980 the state sold 67 percent of its timber by volume through noncompetitive sales. An earlier study found that this was a less effective way to sell timber and generate revenue than auction sales. <FN- Office of the Legislative Auditor, Evaluation of State Timber Sales (St. Paul, 1982), 20-22.> Our analysis shows that:
The volume (in cords) of state-owned timber sold by auction increased from about 44 percent in 1986 to 97 percent in 1996. DNR increased its use of auctions for timber sales on both trust and other state-owned land. Although research indicates that sealed bid auctions generate higher sale prices than oral bid auctions, we found that:
We did not evaluate the Division of Forestry’s overall management of Minnesota’s timber resources, but we think that a study of DNR’s timber management policies and practices, management costs, and timber appraisal and stumpage pricing methods may be timely.
The Division of Minerals administered about 3.4 million acres of school trust mineral rights in 1997. Trust land mineral rights represented 28 percent of the 12.4 million acres of state-owned mineral rights. In addition to iron ore and taconite leases, which have dominated Minnesota’s mining activities, DNR administers metallic minerals, peat, and industrial mineral leases. We found that:
School trust land accounted for about 11,300 acres (35 percent) of the 31,837 acres of leased state mineral rights in 1997. Mineral revenues from school trust land generated about $1.6 million, or 22 percent of total revenues from mineral leases on state-owned land. The cost of mineral management on school trust land is financed with a General Fund appropriation, not from trust land mineral revenues.
The exact nature and location of Minnesota’s mineral resources is unknown, so it is not possible to develop an estimate of the value of those resources. Without knowing the estimated value of the minerals, we are not able to calculate a return on asset value for DNR’s mineral leasing activity on school trust land.
School trust land is located in state parks, wildlife management areas, scientific and natural areas (SNAs), and other DNR units that manage land to provide recreational opportunities and protect critical habitats. Management policies of these units have the potential to restrict or prohibit economic activities on trust land. Both the Minnesota Legislature and DNR have generally recognized the need to compensate the PSF for the lost revenue potential of this land.
In the past, DNR has compensated the trust for lands that could not generate income by purchasing trust land, exchanging trust land for other state-owned land, or paying lease fees for the use of trust land. For instance, in 1992 DNR exchanged over 5,300 acres of trust land in state parks valued at $1.1 million for other state-owned land of similar value. DNR has also used some of its land acquisition money to acquire trust land in SNAs and wildlife management areas. In spite of these efforts, we estimate that:
Table 2 summarizes the distribution of these acres, representing about 6 percent of all school trust land. In addition, between 85,000 and 95,000 acres of trust land are located in wildlife management areas (WMAs), which may limit the revenue generating potential of the land. While DNR policies acknowledge the idea of compensating the PSF when revenues are diminished, we found that:
Since the early 1990s, the Legislature and DNR have added about 550 acres of school trust land to state parks without compensating the trust. Legislation creating the peatland SNAs in 1991 specifically required the Commissioner of DNR to acquire the trust land in these areas. However, DNR does not have any immediate plans to remove the remaining acres of trust land from state parks, SNAs, or other management units. In past years, DNR has given higher priority to acquiring private land in imminent danger of development. Since school trust land is already state-owned and administered, it is not in danger of being developed.
Our analysis also shows that:
In addition to the sale of over 1,000 lakeshore lots and the exchange of trust land in state parks in the late 1980s and early 1990s, there were 39 land sales, 19 land condemnations, and 29 land exchanges involving school trust land between 1987 and 1997.
DNR is faced with a dilemma as it tries to balance its fiduciary responsibilities to the PSF with its natural resource management responsibilities. Given the choice of using limited capital bonding and land acquisition money to acquire trust land versus private land, the department has chosen the latter. In these instances, the department has emphasized its natural resource responsibilities over its fiduciary responsibilities to trust beneficiaries.
Table 2 also shows that approximately 93,000 acres of school trust land valued at approximately $35 million are located in the Boundary Waters Canoe Area Wilderness (BWCAW). Development activities in the BWCAW are severely restricted; this land does not generate revenue for the trust. Recently state and federal officials have discussed the federal government’s purchase of this land. We recommend that:
Between 1986 and 1995, the state sold 1,060 lakeshore lots on school trust land. These sales generated over $14 million in revenues for the Permanent School Fund, mostly (93 percent) financed with annual payments over 20 years.
DNR currently manages 546 lakeshore lot leases on school trust land. Lakeshore leases receive a great deal of public scrutiny, but they account for a small fraction of all trust land. These lots are located on 76 lakes in 12 counties and account for a total of 426 acres of school trust land. The leased lots had an appraised value of $11.6 million as of January 1, 1997. Table 3 shows that based on these appraised values, the lakeshore leases on trust land generated $319,000 in revenues in 1997, the first year of the three-year phase-in of 1997 annual lease rates. Total revenues will increase to $578,000 in 2000.
Some provisions in the initial lakeshore sale laws of 1986 and 1987 benefited lessees instead of trust beneficiaries, such as allowing the lessee to decide if a leased lakeshore lot was to be sold and to cancel the sale after the appraisal was completed. Laws providing for the sale of lakeshore lots have been repealed. However, some provisions in current lakeshore lease laws (1985, 1990) also benefit lessees. Specifically, the three-year phase-in of increases to lease rates reduces revenues for the trust fund. The five-year cycle for adjusting the lease rates based on appraised value is one year longer than the four-year period used to reevaluate the values of other property, including other lakeshore property, for property tax purposes in Minnesota. DNR has argued that some of the lakeshore leasing and sale provisions were not in the interest of trust beneficiaries.
We also examined whether the state should continue to lease the existing lakeshore lots on school trust land or sell them. The analysis rests on a number of assumptions, the most critical of which involve: (1) the rate at which land values will appreciate; (2) the rate of return earned on investment of lease or sale receipts; and (3) how sale costs will be financed. Depending on what assumptions are used, the results of our analysis could support either the continued leasing or the sale of the lakeshore lots. For example, the higher the assumed rate of land value appreciation, the less attractive the option of selling the lots becomes. In considering this issue, policy makers should carefully examine the assumptions used to estimate the costs and benefits of leasing versus selling. Our analysis does not lead to a definitive conclusion, suggesting that any decision about whether to continue leasing or to sell lakeshore lots should not rest on economic analysis alone.
The Minnesota Legislature established the Permanent School Fund Advisory Committee (PSFAC) in 1982 to review DNR land management policies, advise DNR on the management of trust land, and recommend necessary changes in policy and implementation. <FN Minn. Stat. §124.078.>The advisory committee consists of the chairs of the House Education and Ways and Means committees; the Senate Finance and Children, Families, and Learning committees; the Commissioner of Children, Families, and Learning; and two superintendents, one from a nonmetropolitan school district and one from a metropolitan area school district. Our review of the committee and its activities has led us to conclude that:
The PSFAC has met irregularly, usually at the call of DNR. Between 1987 and 1997, the committee met 11 times; between December 1991 and April 1994, the committee did not meet. The committee has focused most of its attention on the leasing and sale of lakeshore lots, the state park land exchange program, and forestry management costs. A legislative staff member who has other significant responsibilities assists the committee.
Through PSFAC is partly composed of legislators who are chairs of major education and finance committees, revenues from the PSF are a small proportion of education finance. Therefore, it is difficult for school trust land issues to capture the attention of these policy makers consistently.
One result of Minnesota’s oversight structure is that no single agency or entity has been responsible for compiling and presenting comprehensive information related to both the school trust land and PSF investments. To address these concerns, we recommend that:
We believe that the Legislature should add the Commissioner of Finance to the PSFAC to serve as chair of the committee and to be responsible for calling regular meetings. The Commissioner of Finance could add financial expertise, a statewide perspective, continuity, and another voice for the interests of the trust to the committee. The committee would retain its current advisory responsibilities of reviewing DNR policies on trust land management, providing advice and guidance to the department, and making recommendations for changes in policy and implementation when necessary. We also recommend that:
Staff support for the PSFAC could assist with the regular review of land management policies and practices and development of a comprehensive annual financial statement on land management proceeds, management costs, deposits to the PSF, and distributions from the PSF. We also think that:
Of the 13 states we examined, Minnesota is one of only two states that use the same structure and staff to manage both trust and other state-owned land. In most other states independent agencies or separate divisions within land management agencies are responsible for trust land management. Typically, an independent agency that deals exclusively with trust land issues is likely to be focused on trust goals and beneficiaries. <FN- Jon A. Souder and Sally K. Fairfax, State Trust Lands: History, Management, and Sustainable Use (Lawrence, Kansas: University of Kansas, 1996), 41- 43.> We do not recommend that Minnesota’s DNR reorganize its administration of trust land to be consistent with the organizations in other states. We suggest, however, that:
DNR could assign a specific staff person responsibility for writing a biennial report, training department staff on the nature of trust land, working as liaison with the PSFAC, and monitoring trust land management activities within the department. We also suggest that the staff person be independent of other DNR functions.
In Minnesota, DNR’s trust land management costs are financed with a combination of land management revenues for forestry management and lakeshore leasing/sales activities and General Fund appropriations for minerals management and other land sales. A national study and our survey of other states show that there are three options for financing the costs of trust land management: (1) general fund appropriations; (2) revenues from trust land management activities; and (3) a combination of land management revenues and general fund appropriations.
Of the states we surveyed, we found that:
Most of the other states we examined use revenues from land management activities to finance management costs. However, these states also have independent agencies or divisions responsible for managing school trust land and are able to identify actual management costs. While forest management costs appear to be reasonably allocated in Minnesota, the allocation of mineral management costs is more complicated. These management costs are not associated with the mineral potential of the land and mineral revenues are not necessarily related to management costs or activities. Based on these considerations, we recommend that:
No change should be made in how Minnesota finances the costs of school trust land management at this time.
Regardless of how management costs are financed, it is unlikely to have an impact on how Minnesota finances education.
We contacted representatives from Minnesota counties with land departments to determine if another unit of government could manage Minnesota’s school trust land more cost-effectively than DNR. After reviewing county land department annual financial reports, we concluded that:
It does not appear that county land departments are equipped to provide the land management services currently provided by DNR, such as minerals management and fire protection and suppression. Decentralizing trust land management could further disperse decision making and complicate the state’s ability to provide comprehensive and consistent oversight.
Proceeds from the sale, use, and management of trust land are added to the Permanent School Fund principal. SBI’s investment of the PSF principal is constrained by constitutional, statutory, and political factors. The Constitution requires that the principal of the fund not be spent. Interest and dividend earnings from the investment of the principal must be distributed to school districts each year. Political and budgetary factors dictate the level of income the PSF is expected to generate for the public schools.
Budget constraints during the 1980s and early 1990s led SBI to invest the PSF principal exclusively in fixed-income securities (bonds) in order to generate the maximum current income for public schools. We found that:
However, this fixed-income investment strategy earned less than alternative portfolio strategies incorporating equities (stocks). In addition, we found that:
The Governor, the Legislature, and SBI recognized this trend and took action in 1997, adopting a budget that expected a reduced contribution from PSF investment income to public education. This has allowed SBI to shift assets from bonds to stocks. In July and August 1997, SBI implemented this initiative by purchasing $212 million of Standard and Poor’s 500 indexed stock portfolio. We support this action, which we recommended in a 1991 report, because it will increase the potential of the PSF principal to grow over the long term. <FN- Ofice of the Legislative Auditor, State Investment Performance (St. Paul, 1991).>
We examined school trust funds in other states and found that:
SBI is recommending further modification of statutory and constitutional restrictions on the way the income and dividends from the PSF are handled. Our analysis indicates that the SBI staff recommendations have considerable merit. Eliminating the restriction on how capital gains are treated would allow SBI to even the cash distribution over time and provide predictable levels of income to the schools. Adopting a distribution policy based on a percentage of market value also would allow the fund to keep up with inflation and ensure that future policies will be consistent with the state’s fiduciary responsibility to the PSF. As a result, we recommend:
Another option that could be explored involves using distributions from the Permanent School Fund for special projects within public education instead of offsetting the general fund education appropriation. For example, the annual PSF distribution might be used to finance capital projects or classroom technology improvements in Minnesota’s public school districts. The PSF could be used for education much like the Environmental Trust Fund is used to finance environmental projects. Under this option, the Legislature would decide how to appropriate the PSF distribution, perhaps with assistance from an advisory group or other body.
Using the PSF distribution for special projects could increase the visibility of the fund, generating more interest in how the PSF principal is invested and how school trust lands are managed. If the PSF distributions were used for specific education projects, school districts around the state would probably become more aware of the trust. Of the states we contacted, Wisconsin and Iowa distribute PSF interest earnings to specific programs.
More InformationThe Program Evaluation Division was directed to conduct this study by the Legislative Audit Commission on May, 1997. For a copy of the full report, entitled "School Trust Land", (98-05), 105 pp., published on March 3, 1998, please call 651/296-4708, e-mail Legislative.Auditor@state.mn.us, or write to Office of the Legislative Auditor, 658 Cedar St., St. Paul, MN 55155.
Staff who worked on this project were Susan Von Mosch (Project Manager), Tom Walstrom and Mary Jackson.
