The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.
M arch 1991 * •iBRARY SEP 27 2001 THE FEDERAL A RESERVE J a RANK of O r ST. IX)l IS ----------------------------------------------------------------------------*— ■ • The Good Job-Bad Job Controversy Involving Services * ■ Home Equity Loans: What Goes Up . . : x '• . ‘ <’ > l) ;lT iyfr;.;: ifl -11 ‘ II ..................... * _; v ; THE EIGHTH FEDERAL RESERVE DISTRICT CONTENTS NTENTS______________________________________________________________________________________ Agriculture District Forest Industry: Growing to New Heights? ............................................................................................. 1 Business Are District Services Jobs Bad Jobs?....................... ................................................................................................. 5 Banking and Finance Home Equity Loans: Flexible Enough to Withstand a Real Estate Downturn?....................................... . .10 Statistics ....................................................................................... ................. .............................................. 14 Pieces of Eight—An Economic Perspective on the 8th District is a quarterly summary of agricultural, banking and business conditions in the Eighth Federal Reserve District. Single subscriptions are available free of charge by writing: Research and Public Information Department, Federal Reserve Bank of St. Louis, Post Office Box 442, St. Louis, MO 63166. The views expressed are not necessarily official positions of the Federal Reserve System. 1 District Forestry Industry: Growing to New Heights? By Jeffrey D. Karrenbrock David H. Kellx provided research assistance. F rom Tennessee's Appalachian mountains, to Arkansas’ Ouchita mountains and to the Mississip pi Delta in the South, much of the Eighth Federal Reserve District is covered by forests.1 This vast natural resource contributes to a wide variety of employment and recreational opportunities, not to mention the stabilizing influence it has on the en vironment. This article examines numerous aspects of the forestry industry in the Eighth District. Prior to identifying the importance of this industry and the potential consequences of environmental re straints, however, the District forest itself is described. The District Forest Forests account for a relatively large portion of the District’s total land area. In 1987, about 44 percent of the District was classified as timberland, that is, land which is capable of crops of industrial wood and not withdrawn from timber utilization by statute or administrative regulation. In the United States, timberland accounts for only about 21 per cent of the total land area. As shown in table 1, the percentage of timberland in Missouri, the Dis trict state with the smallest share, still exceeds the national percentage. Both private and public interests own timber land. Almost 90 percent of the District’s timberland is owned by private entities, while 72 percent of U.S. timberland is held by private entities. The forestry industry is an important private owner in the District, holding 25 percent of the total timber land in Arkansas and 17 percent in Mississippi. Within the District. Arkansas and Missouri have the largest percentage of timberland held by the public, standing at 18 percent and 14 percent, respectively, in 1987. The District’s forests are composed of a mix ture of hardwoods and softwoods. Hardwoods are usually broad-leaved, deciduous trees. Some com mon varieties include oak, hickory, birch, maple, ash, walnut and cherry. Softwoods are usually evergreens, having needles or scalelike leaves. Some examples include longleaf, loblolly and yellow pines. The District’s forests consist primarily of hardwoods, with larger shares of softwoods in the southernmost District states. As shown in table 1, hardwoods account for more than 90 percent of the total growing stock in Kentucky and Missouri and more than 80 percent in Tennessee, while the forests in Arkansas and Mississippi are more even ly split between hardwoods and softwoods.2 On a Districtwide basis, hardwoods compose 70 percent of the growing stock, which compares to only 40 percent of the U.S. forest. District states contain about 17 percent of the nation’s growing stock of hardwood trees. In terms of the total volume of hardwoods and softwoods combined, Arkansas and Mississippi each have more than 19 billion cubic feet of growing stock, more than twice as much as the lowest state, Missouri. Uses of Harvested Timber Harvested timber can be classified into five categories, sawlogs, pulpwood, veneer logs, fuel wood and other products. Sawlogs are used for making lumber and pulpwood is used for making paper products. Veneer logs are used for making veneer finishes and plywoods. Fuelwood is used primarily for home heating. The “ other products” category includes items such as cooperage, pilings, poles, posts, shingles, charcoal and export logs. All District states, except Missouri, (plus Alabama, Louisiana, Texas and Oklahoma) are in the U.S. Forest Service’s “ South Central” region. In the South Central region, sawlogs and pulpwood accounted for 40 percent and 38 percent of soft wood round wood products harvested.3 In the hard wood category, pulpwood, sawlogs and fuelwood accounted for 41 percent, 35 percent and 21 per cent, respectively, of the total roundwood harvest. Growth-to-Removal Ratios The forestry industry is often concerned with how timber supplies are changing over time. Growth-to-removal ratios indicate whether timber removal rates are faster or slower than timber growth. A ratio of less than one would indicate that timber resources are being harvested at a faster rate than that at which they are growing. For the United States as well as all District states, the overall growth-to-removal rate of growing stock timber (softwoods and hardwoods) is well above one, as shown in table 1. Kentucky and 2 Selected Statistics of the Forest Industry U.S. Eighth District AR KY MS MO TN T im b e rla n d as a P ercent of T o ta l* Land A rea 21 44 50 47 55 27 49 P ercent of T im b e rla n d O w ned b y* P ublic P rivate F orest In d u stry 28 72 15 12 88 13 18 82 25 8 93 2 10 90 17 14 86 2 11 89 10 P ercent o f G ro w in g S to ck* H a rd w o o d s S o ftw o o d s 40 60 70 30 55 45 92 8 51 49 92 8 81 19 V o lu m e of G ro w in g S to ck* H a rd w o o d s (m .c.f.)1 S oftw o o d s (m .c.f.) 305054 450881 53140 22753 10655 8586 13500 1110 10069 9746 7334 601 11582 2710 G ro w in g S to ck G row th-toR em oval R a tio 2 All T im b e r H ardw oods S oftw oods 1.36 1.89 1.13 2.18 2.16 2.44 1.23 1.77 .92 2.12 2.15 1.71 1.24 1.81 .98 2.29 2.17 4.08 2.91 3.22 1.91 V alue of T im b e r P ro d u cts3 (m illio n dolla rs) V alue o f A ll O th e r C rops (m illio n d o lla rs) — ... 413 ... 593 ... 265 ... ... 1658 ... 1253 ... 1157 P e rce n t of T otal O utput - 1986 F ore stry P ro d u cts Indu stry L u m b e r & W ood P roducts F u rn itu re & F ixtures P aper & A llie d P roducts 1.7 0.6 0.3 0.8 2.7 0.9 0.6 1.2 4.9 1.9 0.6 2.4 1.4 0.5 0.2 0.7 5.0 2.4 1.5 1.1 1.5 0.3 0.4 0.8 3.0 0.7 0.8 1.5 E m p lo ym e n t (1,000) - 1989 F orestry P ro d u cts Industry Lu m b e r & W ood P roducts F u rn itu re & F ixtu re s P aper & A llie d P roducts 1981.0 757.5 526.4 697.1 237.1 90.8 80.1 66.2 44.1 20.6 9.9 13.6 25.4 11.5 4.5 9.4 59.7 25.3 25.8 8.6 38.0 12.3 12.2 13.5 69.9 21.1 27.7 21.1 SOURCES: Timber data are from various U.S. Forest Service publications. Output figures were derived from U.S. Department of Commerce, Bureau of Economic Analysis. Employment figures are from the U.S. Department of Labor, Bureau of Labor Statistics. 1m.c.f. = million cubic feet M issouri’s statistics are preliminary. Years used to calculate ratios vary across states. 31984 data *1987 data Missouri, in fact, exhibit annual average growing stock growth in excess of twice the amount of removals. For the United States and all District states, except Missouri, softwoods are being harvested at a relatively more rapid pace, compared with their growth rates, than are hardwoods. In fact, the growth-to-removal rate of softwoods is less than one in Arkansas and Mississippi, with most of the harvesting in excess of growth occurring on timberland owned by the forestry industry. Softwood growth-to-removal ratios of less than one in Arkansas and Mississippi may pose a chal lenge for the industry in these states in the future. The U.S. Forest Service projects increasing de mand for softwood products through the year 2040. Whether the forestry industry in these states will be able to maintain their share of the market for 3 this growing demand is questionable. Industryowned forest land is already intensively managed and provides a disproportionate share of softwood removals. For example, a study by the Arkansas Agricultural Experiment Station indicated that in 1985, the forestry industry in Arkansas held 27 percent of the forest land in the state, but accounted for more than 50 percent of the annual softwood supply.4 Policy disputes and legal battles have cur tailed the supply of softwood from Arkansas’ public forests. This implies that any significant increase in softwood supplies must come from the private, non-industrial sector. The above study notes, however, that the incentives for the private sector to invest heavily in future supply are not strong. The expectation of future returns based on current or appreciated prices is not high. Thus, this im plies that relative softwood lumber supplies in some regions will decline in the future. Value of Harvested Timber Products Although often ignored as an agricultural crop, the value of timber production is large relative to the value of all other crops in some states. Estimates by the U.S. Forest Service, shown in table 1, in dicate that the value of harvested timber products in Mississippi was nearly half as large as the value of all other crops grown in the state. In Arkansas and Tennessee, the value of harvested timber pro ducts was about a quarter of the value of all other agricultural crops. Forestry Products Industry and the District Economy The forestry products industry consists of manufacturers of lumber and wood products, fur niture and fixtures, and paper and allied products. The lumber and wood products sector includes out put from logging camps, merchant sawmills, lath mills, shingle mills, plywood mills and veneer mills, among other types of producers. The fur niture and fixtures sector includes output from manufacturers of household and office furniture made of wood.5 The paper and allied products sec tor includes output from the manufacturers of pulp from wood, paper, paperboard, paper bags, paper boxes and envelopes. The forestry products industry accounts for a relatively small portion of the District’s output and employment. These sectors accounted for 2.7 per cent of the District’s output in 1986. As shown in table 1, the relative importance of these industries varies among District states, being most important in Arkansas and Mississippi. Although these in dustries account for a relatively small portion of each District state’s output, they are relatively more important in the District’s economy than they are in the nation’s economy. These three sectors combined accounted for 1.7 percent of the nation’s total output in 1986. In terms of District employment, the forestry products industry accounted for 237,100 jobs in 1989. This was about 3.5 percent of 1989 District non-agricultural employment. Among District states, the forestry products industry employed the most people in Tennessee, followed by Mississippi and Arkansas. In addition to having the largest number of people employed in the forestry products indus try, Tennessee also experienced the largest absolute employment growth (12,000 jobs) in the industry between 1980 and 1989. Part of this growth may be attributable to Memphis’ growth as a national distribution center. As distributional activities in crease, the output and employment of area pro ducers of shipping and packing materials, such as pallets and boxes, also expands. Issues for the Future Future changes in demand and supply condi tions will affect the level of output in the forestry industry. Demand for wood products will likely grow slowly for the foreseeable future.6 The sup ply of wood products, however, is more uncertain. From now until about 2010, the volume of mer chantable softwood sawtimber available in the U.S. is predicted to be inadequate to meet expected demands. Part of this shortfall will be due to restraints placed on timber harvesting because of environmental concerns. Thus, with demand in creasing at a faster rate than supply, wood product prices would increase. On the demand side, long-term construction needs are expected to boost lumber consumption by 23 percent from current levels by the year 2040. Consumption of paper and allied products is expected to more than double its 1986 level by 2040. In terms of wood fiber equivalents, demand for softwood timber is expected to grow 35 per cent and for hardwood, 79 percent. This bodes well for the District’s timber industry as the ma jority of its timber resources are hardwoods. On the supply side, attempts to maintain the habitat of threatened or endangered wildlife and plants may require restricted logging activity and thus lower rates of increase or lower overall sup plies of timber. Many of the current environmental debates focus on issues that will have the largest impact on softwood forests, especially those under control of the national and/or state governments. The spotted owl controversy, for example, largely affects softwood forests. Some restrictions, how ever, also affect the logging of hardwoods. Private forests can also be affected by these restrictions. If we assume that softwood logging will become more restricted, such actions would decrease the supply of softwood timber and, assuming stable or increasing demand for softwood timber, would im ply higher softwood timber prices. Higher soft wood prices would, first, encourage greater soft wood imports. (In 1987, the United States im ported about 23 percent of its total timber con sumption.) Most of the United States’ softwood imports currently come from Canada and any addi tional supplies would also come from that country. However, Canadian forests are also under en vironmental pressures, which are expected to result in lower absolute levels or decreasing rates of in crease in softwood timber production. Thus, the ability of Canada to continue to meet much of the United States’ excess demand for softwood timber is questionable. Higher wood prices would also encourage more recycling of wood and paper products. Cur rently, 21 percent of paper and paperboard produc tion uses wastepaper. If softwood prices rise, then the relative cost of using recycled wastepaper will fall and recycling will expand. The extent to which the percent of paper and paperboard production us ing wastepaper will expand is uncertain. The U.S. Forest Service, however, estimates that by 2040, 26 percent of paper and paperboard production will use recycled wastepaper. Regardless of the amount of increase, it will help to mitigate any potential shortfall in wood supplies. Higher softwood prices would also encourage substitution among types of building material and within types of wood materials. As softwood timber product prices increase, other materials, such as plastic, aluminum and steel, become relatively more attractive for use in construction projects. Thus, we would expect to see an increase in the use of these products at the expense of soft wood products. The higher softwood prices would also encourage the substitution of hardwood pro FOOTNOTES 1The Eighth Federal Reserve District is shown on the map on the inside front cover of this publication. For this article, however, the entire states of Arkansas, Ken tucky, Mississippi, Missouri and Tennessee are referred to as the Eighth District. 2Growing stock is a classification of timber inventory that in cludes live trees of commercial species meeting specified standards of quality and vigor. Cull trees are excluded. 3Roundwood products are logs, bolts and other round timber generated from harvesting trees for industrial or consumer use. 4Much of the information in this paragraph is taken from Kleunder, R. A., E.W. McCoy and J.K. Easterling. The ducts for softwood products. Given the District’s relative abundance of hardwoods, the extent to which District states may gain from this substitution effect would depend in part on the technology available for substitution among hardwood and softwood inputs. Hardwoods can be interchanged with softwoods in some pro duction processes. The more easily the two wood types are substituted, the larger hardwood produc tion response we would expect to see. The amount of increase in hardwood timber harvest, however, may be limited because non-industrial private owners may be unwilling to harvest hardwood timber, instead, preserving it for other purposes, such as for viewing pleasure. The extent of ex panded production may also be limited because hardwood forests are often located in terrain that is difficult and expensive to access. In short, U.S. timber supplies will come under increased pressure in the next several decades as timber demand expands and timber harvesting re strictions are likely to become more prevalent. Whether or not this increased excess demand can be met by larger imports or increased harvesting from non-industrial private forests is questionable. Thus, wood product prices are likely to rise, and some substitution away from wood products can be expected. Summary The timber industry is a key component of the District’s agricultural economy. Whether the Dis trict’s timber industry will benefit from current en vironmental issues that largely restrict softwood harvesting will depend on new technology that in creases the substitutability between hardwood and softwood, the willingness of non-industrial private timber owners to harvest their resource, and the ability of hardwood product manufacturers to devel op and market new products, among other things. Of course, new restrictions on harvesting of hard wood timbers may also arise, limiting potential District economic gains. Arkansas Forest Products Industry, Arkansas Agricultural Experiment Station, Bulletin 908, January 1988. 5Data on this sector also includes employment and out put of metal furniture and fixtures, which obviously is not part of the forest products industry. 6Much of the information in this section was taken from the USDA Forest Service’s The Forest Service Program for Forest and Rangeland Resources, May 1990 and the American Forest Resource Alliance’s The State of Timber Supply - Is the Nation Appropriately Positioned for the 21st Century?, December 12, 1990. 5 by Thomas B. Mandelbaum Thomas A. Pollmann provided research assistance. in service-producing sectors are often viewed as “ bad jobs,” especially when compared with those in manufacturing. This issue is of con cern in the Eighth Federal Reserve District, where the number of jobs in the “ other services” sector rose by more than 572,000 between 1979 and 1990, while manufacturing employment declined by almost 33,000.1 This job shift, which parallels the national experience, reflects manufacturing’s relatively strong productivity gains, as well as the demand of consumers and businesses for increasing quantities of services. Even though the preceding changes can be viewed as positive developments, this article in vestigates the validity of some of these “ good jobbad job” concerns by examining wage rates and other potentially important characteristics of other services jobs in the Eighth District and the nation. What is a Good Job? Most discussions regarding the quality of jobs center around wages: how much is earned, on average, and how many high- and low-paying jobs are provided. To a large degree this focus is ap propriate: wages generally are considered the pri mary determinant of job desirability. When an in dividual compares two job offers, however, other factors also influence the worker’s decision. For example, job stability, the potential for advance ment, non-wage benefits and the chance of injuries are a few of the characteristics that are potentially important to workers. According to one author, the perfect job is one “ with varied duties, little stress, a product that can be seen, problem solving tasks, recognition from the public, flexible hours, high social status, and security, along with high wages.” 2 Even though one may disagree with this characterization of a perfect job, the reality of the labor market is that few such jobs exist. More importantly, the descrip tion reiterates the point that wages are just one, albeit an important one, of the many characteristics that affect workers’ job satisfaction. Earnings o f Full-Time Workers Studies indicating that earnings in the other services sector are relatively low sometimes ignore the depressing effect on wages of the large propor tion of part-time workers in the other services sec tors. As the table shows, when comparing only full-time workers in the District (those working 35 hours per week or more), the median weekly earn ings in other services was $322, which is virtually identical to the $325 median earnings in manufac turing. Workers in both sectors earned somewhat less than the $338 non-farm median. Thus, while other services earnings tends to be below average, the disparity is not great, and when compared with the manufacturing earnings, the disparity essentially vanishes. These comparisons refer only to earnings of wage and salary workers, thus excluding earnings of self-employed workers, some of whom are high-income professionals in services industries. Also excluded are non-wage components of com pensation, such as insurance and paid leave, which tend to be relatively high in manufacturing. If they were included, manufacturing jobs might compare more favorably with jobs in services. In part, the perception of other services jobs as low-paying may stem from the sector’s heteroge neity. Earnings in some of the District’s other ser vices industries, such as the $200 median weekly earnings in personal services, which includes laun dries and barber shops, are considerably less than in manufacturing. On the other hand, professional services workers earn $353 per week, substantially above the level in manufacturing. The table shows substantial variation among the states in the region. In Missouri and Tennessee the pattern of earnings is generally similar to the District average: earnings in other services are slightly less than those in the state’s manufacturing sector and somewhat less than the median of all non-farm industries in the state. In Arkansas, how ever, there is no substantial difference between earnings in other services, manufacturing and the all-industry median; weekly median earnings were near $300 in each case. Interestingly, in all three states, manufacturing jobs, which are generally perceived as “ good jobs,” paid less than the non-farm median. Manu facturing wages in these states were depressed by the large presence of relatively low-paying non durables industries, such as textiles and apparel production in Tennessee, rubber and plastics pro duction in Arkansas and Tennessee, and food pro cessing in all three states. To the extent that job quality is indicated by median earnings, then, the job quality of other services differs little from manufacturing in these three states. usrness Are District Services Jobs Bad Jobs? 6 Level and Distribution of Weekly Earnings of Full-Time Workers, 198912 Percent distribution2 Median Low Middle High U n ite d S tates T otal non-farm M a n u fa ctu rin g O th e r se rvice s $400 406 387 8 .9 % 6.5 10.7 6 9 .2 % 70.0 68.9 2 1 .9 % 23.5 20.4 Eighth D istrict Total non-farm M a n u fa ctu rin g O th e r services 338 325 322 8.9 5.5 12.9 68.6 73.3 67.1 22.5 21.2 20.0 A rka n sa s T o ta l non-farm M a n u fa ctu rin g O th e r services 302 300 305 6.1 3.2 11.1 70.8 78.0 66.2 23.1 18.8 22.7 K e n tu cky T o ta l non-farm M a n u fa ctu rin g O th e r services 362 394 322 11.6 5.3 17.0 64.6 68.4 64.1 23.8 26.3 18.9 M issouri T otal non-farm M a n u fa ctu rin g O th e r se rvice s 365 355 350 10.1 5.9 12.0 65.9 71.0 65.1 24.0 23.1 22.9 T e n n e sse e T otal non-farm M a n u fa ctu rin g O th e r services 320 313 310 5.9 4.4 8.7 69.4 76.2 67.8 24.7 19.4 23.5 includes wage and salary workers. Figures based on data from U.S. Bureau of the Census, Current Population Survey, computer tape, 1990. 2"Low” refers to workers with earnings less than half the median for all non-farm workers in the region, “ Middle” are workers with earnings from one-half to one-and-one-half the median and “ High” are workers with earnings greater than one-and-onehalf the median. Kentucky is another story. Other services workers received median earnings of $322, 11 per cent less than Kentucky’s all-industry median and almost 19 percent less than in manufacturing. Ken tucky’s manufacturing sector is characterized by large employment concentrations in several highpaying industries, including the production of pri mary metals, motor vehicle and tobacco products. In addition to these high-wage manufacturing jobs, Kentucky non-farm earnings are boosted by the abundance of mining jobs. The first thing that stands out when comparing the District with the nation are the nation’s con sistently higher median earnings in all categories. Looking beyond these differences, which in part, reflect the lower cost-of-living in the region, it ap pears that U.S. other services jobs pay substantial ly less than U.S. manufacturing jobs. Other ser vices workers received median weekly earnings that were 4.7 percent less than in manufacturing, and were 3.3 percent less than median earnings in all non-farm jobs. More Low-Wage Jobs? Some observers have contended that, com pared with the manufacturing sector, serviceproducing sectors provide relatively few middleincome jobs, but many low-income jobs, perhaps offset by a few very high-paying jobs. Some pro ponents of this view believe that, besides generating many “ bad” low-income jobs, the growth in other services jobs is undesirable because it will lead to a society in which a growing underclass increas ingly falls further behind a richer class. The table provides some evidence supporting the view that the other services sector offers fewer middle- and high-wage jobs and more lower-paying ones. The middle earnings range is defined as weekly earnings from one-half to one-and-one-half of median weekly earnings of non-farm workers in the region; the low and high categories include earnings lower and higher than the middle range. In the Eighth District, 67.1 percent of workers in the other services sector were in the middle earn ings range compared with 73.3 percent for manu facturing. The other services sector included pro portionately twice as many low-wage jobs than did manufacturing (12.9 percent versus 5.5 percent), and slightly fewer high-wage jobs (20 percent ver sus 21.2 percent). Furthermore, all individual industries within the District’s other services sector had a smaller proportion of workers in the middle earnings range and more in the low range than manufacturing. In the personal services industry, more than a third of workers made less than half the District median, while less than 4 percent fell in the high earnings range. The distribution of earnings in the United States and the four states shown in the table was general ly similar to that of the District: the other services sector had more low-wage and fewer middle-wage jobs than in manufacturing. In Arkansas and Ten nessee, however, other services had relatively more high-wage jobs than manufacturing. While this evidence suggests that in 1989, other services included proportionately fewer middlewage jobs and more low-wage jobs, it does not necessarily imply the employment shift from manufacturing to other services will lead to a twotiered economy. A review of relevant research concluded that while the distribution of earnings in the United States had become somewhat more une qual since the late 1970s, shifts in the nation’s in dustrial mix played only a minor role in the in crease in inequality.3 More fundamentally, while a higher proportion of other services jobs fall in the low earnings range, it is not clear that such jobs are “ bad” jobs. Some analysts suggest that if workers are paid based on their skills and abilities, then the jobs are not necessarily bad. The problem, if any, lies with the skills of the workers rather than with the jobs. Also, these jobs might provide valuable opportuni ties for new workers to acquire work experience and skills. On the other hand, factors other than workers’ skill level, such as an industry’s degree of unioni zation and its capital intensity also are thought to influence wages. In some other services industries, such as personal services, for example, there is relatively little physical capital per worker com pared with manufacturing, which tends to reduce productivity and, therefore, wages. One important difference between manufactur ing and other services relates to how earnings are related to formal educational achievement.4 Con siderable formal education is generally required to get a high-paying job in service occupations. In those other services industries with the highest wages, such as professional services, educational levels are high. In contrast, formal education is not as important in manufacturing, where on-the-job training appears more important. Thus, many manufacturing workers who have lost their jobs, but lack high levels of formal education, may find it difficult to find services jobs with comparable wage rates unless they first gain additional skills through education or training. Are Part-Time Services Jobs Bad? A relatively high proportion of wage and salary employees in other services work part time. In 1989, for instance, 24 percent of the nation’s other services workers were on part-time sched ules, compared with 17.6 percent for all non-farm industries and just 5.6 percent in manufacturing.5 Compared with full-time workers, those who work part time tend to earn lower wages and are less likely to receive pensions, health insurance and other benefits. Thus, the abundance of parttime work in the other services sector has been cited as evidence of low job quality. This is not necessarily the case as many prefer the flexibility and increased time for family and leisure that these jobs allow. For these people, the greater availabili ty of part-time jobs in the other services sector is one of its desirable characteristics. To the extent that part-time schedules are un wanted by workers, however, the high proportion of such jobs in other services is an undesirable in dustry characteristic. For the most part, this does not appear to be the case: of all U.S. part-time workers in other services, just one in five worked part time involuntarily in 1989. This compares with 23 percent of part-time workers in all in dustries and 41 percent of part-time manufacturing workers.6 Comparative Stability In contrast to many non-wage job characteris tics, employment stability is readily measurable. As figure 1 clearly shows, employment in other services is much more stable than in manufacturing in the Eighth District, as is also true nationally. In times of national recession (shaded in the figure), manufacturing output and employment tend to decline sharply as consumers postpone their pur chases, especially of durable goods like cars and appliances. While other services employment is also affected by recession — note its deceleration 8 Figure 1 Eighth District Manufacturing and Other Services Employment Percent Annual rate of change Other Services V Aw, 4 I ___ I I 72 74 l 76 i I 78 in the mid-1970s and early 1980s — it is less sen sitive than manufacturing. Compared with manu factured goods, consumers are less likely to put off the purchase of many kinds of services, like medical procedures or haircuts. Perilous Conditions? Another measurable characteristic is hazards of jobs in various industries. Hazardous jobs, which often involve working with dangerous equipment or materials, are widely viewed as less desirable than those which do not entail such risks. Accor ding to one industry hazard indicator — the in dustry’s incidence of occupational injury and il lness — jobs in other services are much less hazar dous than average. In the other services sector, 51.2 workdays were lost due to injury and illness per 100 full-time workers in 1989.7 This rate com pares with 78.7 lost workdays for all private sector \ acturing \ I l Ml * I \ 1 Manu V 1970 lfk/VAV r' < Ua I 80 ■ I 82 I _1___ L . 84 86 i 88 i 1990 workers and 113 lost in manufacturing. While there is undoubtedly wide variation among dif ferent types of occupations within the other ser vices sector, these figures suggest that, in general, such jobs are relatively less hazardous than in manufacturing or other sectors. Conclusion Like most simple questions, the one posed in the title of this article has no simple answer. Most fundamentally, it is not clear what constitutes a bad job. If one considers the median pay levels among full-time wage and salary workers in the Eighth District, workers in the other services sectors earn somewhat less than the non-farm average, but es sentially the same as those in manufacturing. The median earnings data, however, hide the fact that the District’s other services sector has relatively larger shares of low-paying jobs and smaller shares 9 of middle-earnings jobs than manufacturing. How these facts should be interpreted, however, is a source of controversy. Despite the relatively large number of lowwage jobs compared with manufacturing, workers in the other services sector have the security of working in a more stable industry and also ex perience a lower incidence of injuries and illnesses. The lower wages in some other services jobs might FOOTNOTES 1The other services sector is composed of health, business, personal, professional, repair, legal and miscellaneous services. For a more complete descrip tion see “ District Services: What They Are and Why They Have Grown” by Thomas B. Maridelbaum in Pieces of Eight (December 1990). Data for Arkansas, Kentucky, Missouri and Tennessee are used to repre sent the Eighth District. 2Neal H. Rosenthal, “ More Than Wages at Issue in Job Quality Debate,” Monthly Labor Review (December 1989), p. 7. 3Gary W. Loveman and Chris Tilly, “ Good Jobs or Bad Jobs: What Does the Evidence Say?” Federal Reserve be offset, in part, by these positive industry attri butes as well as by other job characteristics that are less easily measured, but nevertheless con tribute to a worker’s evaluation of job quality. Finally, the other services sector had a high pro portion of part-time workers; however, since most of these employees worked part time voluntarily, the abundance of part-time jobs is not in itself undesirable. Bank of Boston, New England Economic Review (January/February 1988), p. 46-65. 4See John R. Swinton, “ Service-Sector Wages: the Im portance of Education,” Federal Reserve Bank of Cleveland, Economic Commentary (December 15, 1988). 5U.S. Department of Labor, Bureau of Labor Statistics, Employment and Earnings (January 1990), p. 199. 6lbid. 7See U.S. Department of Labor, Bureau of Labor Statistics, Monthly Labor Review (December 1990), pp. 109-10. g &Finance 10 Home Equity Loans: Flexible Enough to Withstand a Real Estate Downturn? by Michelle A. Clark Thomas A. Pollmann provided research assistance. just one decade, home equity loans have turned into one of the most successful new pro ducts ever offered by U.S. financial institutions. Buoyed by rapid home price appreciation in many parts of the country, home equity loan growth sur passed that of most other categories of loans in the 1980s. Although home prices rose more slowly in the Eighth District than in New England and California, District banks also experienced strong growth in home equity lending in the 1980s. Re cent trends in the home equity loan market and reasons for the popularity of these products among consumers and bankers are explored below. How a Home Equity Loan Works Most people are familiar with the oldest type of home equity loan, also known as a second mort gage. Traditional home equity loans (called closedend home equity loans) are paid out in full at the time of origination and usually require repayment of interest and principal in equal monthly install ments over a fixed time period. Closed-end loans can be used for a variety of purposes; however, they typically have been used for large, one-time expenses, such as a home improvement project. Although closed-end home equity loans are still of fered by many institutions, their growth since the mid-1980s has been eclipsed by that of the more flexible home equity line of credit. A home equity line of credit (HELOC) is an open-end revolving account secured by residential equity, and works more like a credit card account than a mortgage loan. A HELOC account allows discretionary borrowing up to the amount of the credit line, and can usually be accessed through a special checking account or credit card. Most HELOC accounts feature variable interest rates, generally the prime rate plus 2 percentage points. HELOC accounts are most often used to finance home improvement projects, but they are also used for debt consolidation, medical expenses, tuition payments or the purchase of a new car or home appliance. In 1980, less than 1 percent of all commercial banks and thrifts offered HELOC accounts; today about 80 percent of commercial banks and 65 per cent of thrifts offer these loans, with commercial banks dominating in the number and volume of outstanding credit lines. While a number of regulatory and economic factors contributed to the tremendous growth in this product during the 1980s, the major explanatory factor appears to be the substantial increase in home prices from the late 1970s through most of the 1980s. These home price increases, especially in New England, the Mid-Atlantic states and the West Coast, dramatically improved the equity positions of most households, regardless of how long they had owned their homes. An example will show how this can be true. Say a homeowner with a house valued at $100,000 has $30,000 in equity invested in that house and a $70,000 mortgage. Suppose five years later this homeowner could sell this house for $150,000. Assuming this apprecia tion is permanent, the homeowner (and her credi tors) would view this unrealized profit as equity or household savings. Home equity loans allow the homeowner to mobilize this household wealth without selling the home; the homeowner borrows against the equity in her home, which, because of rapid home price appreciation, may exceed the original mortgage. Data collected on the character istics of home equity loans bear this home price appreciation story out: home equity lending is most prevalent in states where home values are highest and have appreciated the most. Home Equity Loan Trends It is estimated that during the second half of 1988, 6.5 million households (or 11 percent of all U.S. households) had home equity loans, with the proportion holding closed-end and HELOC ac counts roughly equal. This data does not reflect, however, the substantial gain in popularity of HELOC accounts over closed-end home equity loans since the mid-1980s. In 1988, 63 percent of all home equity loan originations were HELOC ac counts versus a 37 percent share for closed-end loans. Because growth in HELOC accounts has dominated that of closed-end loans during the last several years and because HELOC accounts have some features unique among loan products, they will be the focus of the remainder of this article. Prior to December 1987 for commercial banks and December 1988 for savings institutions, HELOC accounts were included in total residential mort 11 gages when these institutions reported loan data to their supervisory agencies. These old reporting practices make it difficult to pinpoint total growth in HELOC accounts during the 1980s.1 Since the data have been reported separately, however, it is clear that HELOC account growth has far sur passed that of most other types of bank loans, and offering these loans has allowed many institutions to substantially expand their retail banking activity. As illustrated in table 1, HELOC accounts make up a small yet rapidly growing share of total loans at U.S. banks. At the end of September 1990, home equity lines of credit accounted for 3.05 percent of total loans nationally and 2.13 per cent of total loans in the District, both up more than 30 percent from year-end 1988. For various reasons, but mostly because home prices have not appreciated much in the Midwest during the last few years, HELOC accounts make up a smaller proportion of District bank portfolios than they do nationally. HELOC accounts are most prevalent in the Northeast, where the median price of an ex isting home rose 122 percent between 1982 and 1990. In the Midwest, home prices rose a more modest 34 percent during the last eight years, slightly less than the national rise of 41 percent. Led by banks in Illinois, Missouri and Tennessee, District banks have experienced slightly faster average growth in HELOC accounts since year-end 1988 than U.S. banks overall; yet, only in Illinois and Tennessee are the September 1990 averages close to the national average. In contrast, Arkansas banks have much lower HELOC shares than District and U.S. banks, with a September 1990 share of just 0.33 percent, less than one-quarter the District average. Why So Popular? Climbing home values alone do not explain why a homeowner would choose a HELOC ac count versus another type of consumer loan. Con sumers have cited two characteristics of HELOC accounts which largely explain their popularity: convenience and the continued federal tax deduc tion for mortgage interest expense. When banks began heavily promoting HELOC accounts in 1986, most waived closing fees and did not charge any fees to maintain the credit lines. As a result, many households established accounts in anticipa tion of large future expenses, in much the same way businesses apply for standby letters of credit. The idea of being able to tap into a line of credit, at any time, for any reason, and for any amount up to the maximum, appealed to many consumers. So too did the tax break HELOC bor rowers receive. The 1986 Tax Reform Act pro vided for the gradual elimination of the deduction Table 1 Home Equity Lines of Credit as a Percent of Total Loans U nited S ta te s 1 Eighth D istrict A rkansas Illinois Indiana K e n tu cky M ississip p i M issouri T e n nessee 9/30/90 12/31/89 12/31/88 3 .0 5 % 2.13 2.78% 1.86 2 .34% 1.56 0.33 2.78 1.92 2.08 1.03 2.76 2.78 0.33 2.25 1.84 1.94 0.89 2.26 2.51 0.29 1.59 1.54 1.68 0.78 1.93 1.97 includes only U.S. banks with assets of less than $10 billion. NOTE: State data are for whole state, not just the portion located within the Eighth District. SOURCE: FFIEC Reports of Condition and Income for In sured Commercial Banks, 1988-90 for most types of consumer interest expense; how ever, the deduction for interest on loans secured by residential property, which include both types of home equity loans, was not changed. Because HELOC accounts have few restric tions on use and because they have features com mon to consumer installment loans and credit card accounts, often at much lower interest rates, many borrowers have substituted HELOC account draw downs for traditional consumer loans. Consumer surveys indicate that many HELOC account holders are foregoing auto loans, student loans and credit card purchases in favor of borrowing against their home equity accounts. One way to assess the extent of this substitution is to examine the ratio of HELOC accounts to consumer loans. For U.S. banks with assets of less than $10 billion, this ratio rose from 10.45 percent at year-end 1988 to 13.54 percent in September 1990, and for District banks, from 7.38 percent to 10.21 percent. These numbers support the assertion that borrowers are using home equity loans to finance a variety of consumer purchases. The Banker’s View The popularity of HELOC accounts extends to bankers too, although for different reasons. From the banker’s point of view, the home equity credit line has a number of attractive properties: it is a high-yielding asset and one that is secured by collateral (residential property), unlike many 12 Figure 1 Unused Home Equity Lines of Credit as a Percent of Home Equity Lines of Credit Balances Outstanding, 1990 consumer loans. In addition, HELOC accounts are attractive to bankers because they are easily pack aged with other banking products, allowing banks to expand total retail services. HELOC accounts are only profitable to banks, however, if they are used. Since there was often no cost to borrowers to set up HELOC accounts, no penalty for not us ing them and no minimum drawdown, many bankers did not see a return on their marketing and start-up costs for a number of years. Based on consumer and banking surveys, it appears more people who have opened HELOC ac counts during the last several years are currently using them or are increasing their outstanding balances than in previous years. This increased usage is partially attributed to the elimination of the tax deduction for consumer interest expense, which is almost fully phased in, and to the lapsing of many accounts opened several years ago that have never been used. Maintaining low levels of unused home equity accounts relative to balances outstanding is a key factor in making HELOC accounts profitable for the bank. Unused home equity lines of credit, an off-balance sheet item, represent a commitment on the part of the bank to lend at a future date. In recognition of that commitment, banks are now re quired to hold capital against a portion of those unused lines of credit as well as credit line bal ances outstanding. An unused credit line represents a burden to a bank because there is an “ expense” (the capital requirement) and no return; the lower the ratio of unused credit lines to outstanding HELOC balances, all else equal, the more pro fitable the product is to the bank. The ratios of unused credit lines to HELOC balances outstanding at U.S. and District banks in 1990 are illustrated in the figure. District banks, largely because of Tennessee banks, have a greater ratio of unused HELOC accounts to outstanding balances than U.S. banks overall. Tennessee banks, on average, have more commitments to make HELOC loans than actual loans on the books, an 13 indicator that this product is not as profitable in Tennessee as elsewhere in the District and the United States. The relative safeness of HELOC accounts for banks shows up in nonperforming loan statistics. As illustrated in table 2, delinquency rates on HELOC accounts are substantially below that of traditional home equity loans and closed-end con sumer loans, both nationally and throughout the District. Many analysts attribute the lower delin quency rate on HELOC accounts to borrower char acteristics; a number of consumer surveys have shown, for example, that the typical HELOC bor rower is in a higher income bracket and is more educated than the average home equity or con sumer loan borrower.2 These characteristics are associated with lower delinquency rates, presum ably because wealthier people are less likely to run into payment problems and more-educated bor rowers are less likely to make ill-informed choices. Despite current low delinquency rates, a num ber of regulatory and structural characteristics of home equity lending present risks to the lender. The Competitive Equality Banking Act of 1987 re quired creditors, including banks, to establish a life-of-loan cap on all adjustable rate mortgages; if interest rates were to rise dramatically, a bank may incur losses on capped loan products, includ ing HELOC accounts (called interest rate risk). Unlike regular mortgage loans, HELOC accounts are difficult to securitize (sell in the secondary market) because of the inconsistency of their in terest and credit risk characteristics, the complexity of their collateral structures and the uncertainty of payment dates because of their revolving credit nature. The inability to securitize HELOC accounts means banks have to hold capital against them and are unable to pass on some of the interest rate and credit risk to secondary market participants. The biggest risk to HELOC lenders, however, is the effect of a decline in housing values. If a decline is steep enough to cause a loss of house hold purchasing power, the bank is subject to the risk of property abandonment and housing debt default. Although any lender of funds backed by residential property faces this risk, the treatment of lien priority makes this risk especially significant for HELOC lenders. If a homeowner defaults on his mortgage and his HELOC, the bank holding the HELOC is more likely to suffer losses because HELOC accounts are typically secured by junior as opposed to first liens. Given that a number of regions are experiencing declines in home prices FOOTNOTES 1Closed-end home equity loans are still reported with other one- to four-family residential mortgages, making it even more difficult to quantify the total amount of home equity loans outstanding. Table 2 Loans 30 Days or More Past Due as a Percent of Loans Outstanding, Year-End 1989 H ELO C a cco u n ts T ra d itio n a l hom e e q u ity loans C o n su m e r loans— closed-end U nited S tates 0 .7 8 % 1.85% 2.95% A rkansas Illinois Indiana K e n tu cky M ississip p i M issouri T enn e sse e 1.26 0.73 0.59 0.61 0.25 0.94 2.15 1.11 2.35 3.07 2.39 1.49 2.21 2.12 1.81 2.68 3.58 3.06 2.65 3.13 — SOURCE: Consumer Credit Delinquency Bulletin, American Bankers Association and these regions are the ones with the greatest shares of home equity loans, it is likely that banks will experience rising delinquency rates during the next year. Conclusion Home equity loans, and in particular, home equity lines of credit accounts, rank high on the list of significant financial innovations of the 1980s. Their popularity among consumers has generated some large profits and market shares for banks aggressively promoting the products. The flexibility of these loans has been their most popular feature, and continued innovation will un doubtedly increase their appeal. The resiliency of the product, however, rests in its ability to per form well in a time of declining home prices. The current downturn in real estate markets in New England and the Mid-Atlantic states offers the first real test of this product’s staying power. For Eighth District bankers contemplating a jump into the market, what happens on the East Coast during the next 18 months will surely provide food for thought. 2See the American Bankers Association 1990 Home Equity Lines of Credit Report and The Survey Research Center, University of Michigan National Survey of Home Equity Loans, September 1989. 14 Eighth District Business Level Payroll Employment (thousands) United States District Arkansas Little Rock Kentucky Louisville Missouri St. Louis Tennessee Memphis M anufacturing Employment (thousands) United States District Arkansas Kentucky Missouri Tennessee District Nonmanufacturing Employment (thousands) Mining Construction FIRE2 Transportation3 Services Trades Government Real Personal Income4 (billions) United States District Arkansas Kentucky Missouri Tennessee C o m p o u n d e d A n n u a l R a te s o f C h a n q e IV/1990 111/1990IV/1990 IV/1989IV/1990 19901 19891 110,205.0 6,888.2 922.9 250.0 1,473.9 487.3 2,325.4 1,181.2 2,166.1 472.5 - 1 .6 % - 0 .4 2.1 2.8 -0 .9 2.3 0.3 0.6 - 1 .8 5.5 0.9% 0.8 2.5 1.6 1.8 2.1 0.4 -0 .1 -0 .1 2.1 1.8% 1.5 2.8 1.9 2.6 3.0 0.8 0.8 1.1 1.8 2.7% 2.9 3.0 3.0 3.8 4.1 2.2 2.3 3.0 1.5 18,798.0 1,456.9 231.3 282.8 426.7 516.1 - 5 .7 % -3 .5 -2 .4 -3 .4 -5 .2 - 2 .7 - 2 .6 % - 1 .3 1.0 - 0 .8 - 2 .4 -1 .7 1.9% -0 .4 0.6 0.2 -1 .3 -0 .5 0.4% 1.9 1.6 3.4 1.2 1.9 - 1 .6 % -3 .3 0.6 0.7 4.6 0.3 -2 .5 - 0.4% - 2 .0 -0 .1 0.2 2.8 0.7 2.2 - 0.6% 1.4 0.3 0.4 3.4 1.4 2.8 - 4.8% 1.0 0.3 3.5 5.1 3.0 2.2 111/1990 11/1990111/1990 111/1989111/1990 19891 19881 $3,545.3 192.7 25.2 41.5 67.3 57.8 - 1 .3 % - 2 .5 - 4 .6 - 2 .8 - 2 .9 -0 .7 0.7% 0.3 2.0 0.7 - 0 .7 0.3 2.7% 1.9 1.6 2.5 1.8 1.7 3.9% 2.8 2.5 2.8 1.7 4.4 IV/1990 111/1990 5.9% 6.2 7.2 5.9 5.7 4.9 6.2 6.4 5.9 5.2 5.6% 5.8 7.0 6.0 5.5 5.0 6.1 6.4 5.1 4.6 49.0 292.6 338.3 398.3 1,573.4 1,649.2 1,131.8 L e v e ls Unemployment Rate United States District Arkansas Little Rock Kentucky Louisville Missouri St. Louis Tennessee Memphis 1990 5.5% 5.7 6.8 5.8 5.7 5.0 5.7 5.9 5.3 4.7 1989 5.3% 5.8 7.2 6.3 6.2 5.6 5.5 5.5 5.1 4.7 1988 5.5% 6.5 7.7 6.4 7.9 6.3 5.7 5.9 5.8 5.2 Note: All data are seasonally adjusted. On this page only, the sum of data from Arkansas, Kentucky, Missouri and Tennessee is used to represent the District. 1Figures are simple rates of change comparing year-to-year data. 2Finance, Insurance and Real Estate ^Transportation, Communications and Public Utilities 4Annual rate. Data deflated by CPI-U, 1982-84 = 100. 15 U. S. Prices L ev el IV /1 9 9 0 C o m p o u n d e d A n n u a l R a te s o f C h a n q e 111/1990- IV /1 9 8 9 - IV /1 9 9 0 IV /1 9 9 0 19901 19891 Consumer Price Index ( 1982 - 8 4 = 100 ) Nonfood Food 133.4 134.4 7.2% 4.0 6.4% 5.5 5.3% 5.8 4.7% 5.8 144.7 167.0 121.7 -13.4% -13.8 -13.0 - 1 .1 % 1.0 -3 .6 1.6% 6.8 -4 .8 6.6% 6.8 6.6 174.0 187.0 7.2% 6.7 4.8% 5.1 2.7% 3.4 6.4% 4.9 Prices Received by Farmers (1 9 7 7 = 100 ) All Products Livestock Crops Prices Paid by Farmers (1 9 7 7 = 100 ) Production items Other items2 Note: Data not seasonally adjusted except for Consumer Price Index. 1Figures are simple rates of change comparing year-to-year data. 2Other items include farmers’ costs for commodities, services, interest, wages and taxes. Eighth District Banking Changes in Financial Position for the year ending September 30, 1990 (byAsset size) SELECTED ASSETS Securities U .S. Treasury & agency securities Other securities1 Loans & Leases Real estate Commercial2 Consumer Agriculture Loan loss reserve Total Assets SELECTED LIABILITIES Deposits Nontransaction accounts MMDAs $100,000 CDs Demand deposits Other transaction accounts3 Total Liabilities Total Equity Capital Less than $ 1 0 0 m illio n • $ 3 0 0 m illio n - M o re th a n $ 1 0 0 m illion $ 3 0 0 m illio n $1 b illio n $1 b illio n 3.4% 6.1 - 5 .2 2.5 4.4 -4 .1 - 0 .5 11.5 0.3 1.5 1.5% 2.4 -5 .4 6.0 -5 .3 1.8 1.5 1.5 14.7% 20.7 -0 .2 5.3 11.5 -4 .4 -1 .4 15.0 9.0 7.8 8.3% 9.8 3.6 4.5 0.0 8.0 7.7 8.8 17.2% 24.6 0.1 5.6 10.1 -4 .3 12.2 19.9 3.4 7.5 22.2 -4 .5 3.1 15.8 -1 .9 3.7 -5 .3 17.9 4.1 9.2% 11.7 5.4 - 2 .8 -3 .6 12.8 7.7 5.7 7.8% 10.0 17.4 9.0 0.1 9.3 4.4 26.0 Note: All figures are simple rates of change comparing year-to-year data. Data are not seasonally adjusted. 11ncludes state, foreign and other domestic, and equity securities includes banker’s acceptances and nonfinancial commercial paper includes NOW, ATS and telephone and preauthorized transfers 14.2% 16 Performance Ratios (by Asset Size) III/90 EARNINGS AND RETURNS Annualized R eturn on Average Assets Less than $100 million $100 million - $300 million $300 million - $1 billion $1 billion - $10 billion More than $10 billion Agricultural banks Annualized R eturn on Average Equity Less than $100 million $100 million - $300 million $300 million - $1 billion $1 billion - $10 billion More than $10 billion Agricultural banks Net Interest M argin1 Less than $100 million $100 million - $300 million $300 million - $1 billion $1 billion - $10 billion More than $10 billion Agricultural banks ASSET QUALITY2 Nonperforming Loans3 Less than $100 million $100 million - $300 million $300 million - $1 billion $1 billion - $10 billion More than $10 billion Agricultural banks Loan Loss Reserves Less than $100 million $100 million - $300 million $300 million - $1 billion $1 billion - $10 billion More than $10 billion Agricultural banks Net Loan Losses4 Less than $100 million $100 million - $300 million $300 million - $1 billion $1 billion - $10 billion More than $10 billion Agricultural banks Eighth District III/89 United States III/88 lli/90 III/89 ill/88 .83% .91 .77 .57 .47 1.09 .85% 1.00 .88 .81 .06 1.12 .740/0 .85 .68 .75 .91 1.01 8.98% 11.07 10.36 8.57 9.41 11.14 9.20% 12.24 12.23 12.52 1.31 11.40 8.25% 10.85 9.76 11.87 18.85 10.42 4.09% 4.30 4.36 4.17 3.26 4.02 4.29% 4.45 4.38 4.14 3.39 4.14 4.25% 4.25 4.15 4.06 3.30 4.07 1.06% 1.03 1.04 .79 1.11% 1.08 1.04 .58 1.06% 1.01 1.07 .86 1.23 1.21 1.15 11.41% 12.35 13.19 11.82 11.82% 13.00 13.08 8.99 11.56% 12.16 13.46 12.80 — 12.30 3.99% 3.91 3.96 3.75 — 12.11 4.00% 3.98 4.11 3.59 — 11.68 3.91 3.92 3.95% 3.88 4.04 3.72 — 3.83 1.65% 1.85 1.55 1.94 1.76 1.65% 1.72 1.38 2.18 — 1.87 1.820/o 1.72 1.33 2.03 — 2.08 2.01% 2.04 2.50 3.04 4.78 1.89 2.16% 1.95 2.49 2.26 4.87 2.22 2.440/o 2.01 2.19 2.13 5.53 2.70 1.44% 1.49 1.39 1.76 1.47% 1.43 1.42 1.79 1.46% 1.36 1.32 1.82 1.51% 1.49 1.77 2.20 3.38 1.93 1.56% 1.46 1.62 1.80 4.24 2.04 1.63% 1.50 1.64 1.79 4.18 2.08 .35% .42 .58 .99 1.37 .24 .42% .38 .51 .62 .80 .32 .520/0 .44 .56 .74 .77 .47 — — — 1.61 .25% .36 .33 .61 — — 1.74 .23% .32 .34 .59 — 1.75 .27% .32 .29 .83 — — — .16 .19 .25 Note: Agricultural banks are defined as those with 25 percent or more of their total loan portfolio in agriculture loans. 'Interest income less interest expense as a percent of average earning assets 2Asset quality ratios are calculated as a percent of total loans. 3Nonperforming loans include loans past due more than 89 days, nonaccrual, and restructured loans. 4Loan losses are adjusted for recoveries.