2004 Assessment Ratio Report

The Department of Assessments and Taxation appraises real property for the purposes of property taxation. Properties are valued using the three approaches to value generally recognized by the appraisal profession: cost, sales comparison, and (when applicable) income.

In Maryland, all properties are required by law to be physically reviewed once every three years. During the review, the assessor will visit properties to verify property characteristics existing in our current assessment records. Residential property characteristics include type of structure, size, quality and type of construction, condition of structure, and any new improvements. In certain circumstances, neighborhood inspections may be made in place of individual property inspections. Commercial properties are reviewed for type of structure, size, type and quality of construction, condition of structure, current use of the property, any new improvements, types of tenants, and vacancy.

We value over 680,000 properties each year, which requires the use of mass appraisal techniques. While a fee appraiser is concerned with valuing one property at a time, an assessor is valuing whole neighborhoods. To accomplish this, special mass appraisal procedures are used. The assessor will review the data and calculate replacement costs for improvements much like a fee appraiser. The assessor will then review the sales from the area. In Maryland, the local assessment office, except in Baltimore City, receives a copy of all deeds and property sales prices as the deed transferring the property is recorded with the clerk of the court. In Baltimore City, the Department of Public Works does the data entry and provides the data to the Department. In the assessor’s review and analysis of the sales, the assessor will develop land rates, depreciation tables, and sales analysis reports. After completing the analysis, the assessor applies the factors uniformly throughout the neighborhood to value all comparable properties in a uniform manner. Rental rates, vacancy and collection loss, expense ratios and capitalization rates are analyzed, and uniformly applied for comparable income producing properties.

The Department’s work is reviewed by legislative auditors and is often scrutinized by individual property owners. We are continually striving for higher quality in assessment uniformity. Our quality control program begins with the individual assessor and the assessor’s immediate supervisor. As work is completed, each assessor’s supervisor reviews the analysis, makes recommendations, and approves the work. When the assessor completes the revaluation, the supervisor makes a random check using procedural and data editing checks. Following the completion of the revaluation, various computer edits are made to assure good valuation quality.

A measurement of quality is the assessed value/sale price ratio. A ratio is the relationship of two numbers, in this case assessed value and sale price. It measures how closely our values compare to the actual sales prices. The average assessed value/sale price ratio indicates a typical level of value. Because the marketplace is not perfect, there will always be properties that sell for more or less than can be anticipated due to factors such as sales between people unfamiliar with the market, buyers willing to pay extra for a unique property, or escalating values in a competitive seller’s market.

In mass appraisal and assessment ratio studies, we are not only concerned with average assessed value/sale price levels (ratios) but also with the degree of spread (variation) from the typical ratio. The measurement of variation is called the coefficient of dispersion (COD). The lower the COD, the more uniform the assessment level.

In the balance of this report, Section II will give a more detailed explanation of the statistical terms as applied to assessment administration and quality control. Section III explains the International Association of Assessing Officer’s Standard of Performance for ratio studies. Section IV gives an overview of statewide appraisal quality for the most recent valuation of triennial Group 1, performed in December 2003.

SECTION II – RATIO STATISTICS

The purpose of this ratio study is to test the quality of the assessment product. The quality of the assessment product is examined from both an assessment level and assessment uniformity standpoint. Assessment level examines the degree to which the assessments are performed based upon the statutory requirement of full market value. Assessment uniformity measures the degree to which different properties are assessed at equal percentages of their market values. From our most recent valuation, we perform many ratio studies examining neighborhoods, types of structures, age of structures, etc.

We use as a performance gauge several measures of central tendency. A ratio of assessed value to sale price is calculated for each property. The average ratio is the total of all ratios divided by the number of sales. The median is the midpoint of any data listed from lowest to highest. The median ratio is the point where half the ratios fall above and half ratios fall below. The weighted ratio is the total of all assessed values divided by the total of all sale prices. Each measure of central tendency is affected differently by outliers.

In addition to the general level of assessments, we are also concerned with the relative spread or variation that individual ratios fall from the typical. There are two measurements of variability: coefficient of dispersion and coefficient of variation. The coefficient of dispersion is calculated by dividing the average absolute deviation by the median ratio. The average absolute deviation is calculated by subtracting the median ratio from each ratio, adding all the results but ignoring positive and negative signs, and dividing by the number of ratios. Acceptable coefficients of dispersion depend on property type but should typically be 20% or less. Coefficient of variation is calculated by dividing the standard deviation by the mean or average ratio and multiplying by 100. The variance is calculated by subtracting the mean from each ratio, squaring the differences, summing the squared differences, dividing by the total number of ratios less one. The standard deviation is calculated by taking the square root of the variance. The coefficient of dispersion is the preferable measure of variance unless a sample is normally distributed. In a normal distribution situation, coefficient of variation is the preferable measure of variance.

Another statistical measure used to gauge assessment uniformity is the Price Related Differential (PRD). The PRD tests to see if higher or lower valued properties are assessed at the same level. It is calculated by dividing the average ratio by the weighted ratio. Typically, PRDs have an upward bias because higher priced properties are more unique. PRDs should range between 0.98 and 1.03, except for very small samples. For example, a PRD of 1.03 indicates under valuation of high priced properties, while a PRD of .98 shows an under valuation of low priced properties.

Other descriptive statistical methods that may be used to analyze the assessment product are histograms, frequency distributions, and scatter diagrams. Due to the scope of this report, we have not examined them here. For further information on statistics relating to assessments, please refer to the International Association of Assessing Officers’ publication “Improving Real Property Assessment”.

Table I is the Fiscal Year 2005 Real Property Base/Ratio by Subdivision with assessment ratios expressed relative to full value. Table II is a history of weighted assessment ratios converted to full value (100% levels) that allows for comparison between years by adjusting for statutory changes in the assessment level. Table III displays examples of the statistical calculations used in this report.

Tables IV and V show the residential and commercial 2004 Ratio Study data by subdivision at assessed full market value level for the area most recently assessed. Following the ratio study is Table VI of the report detailing issues of assessment and appraisal quality that are summarized in Section IV.

SECTION III – RATIO STUDY STANDARDS VALUES TO SALE PRICES

The International Association of Assessing Officers (IAAO) is a professional organization of assessing officials which provides educational programs, assessment administration standards, and research on appraisal and tax policy issues. IAAO has developed numerous standards and texts on appraisal and assessment administration. Additionally, the organization is a founding member of the national Appraisal Foundation which developed the Uniform Standards of Professional Appraisal Practice (USPAP).

IAAO’s Standard on Ratio Studies was first published in September 1980 and was revised in July of 1999. The Standard is advisory in nature. This Standard provides guidance to those performing ratio studies in the mass appraisal field regarding the design, statistics, performance measures and other issues related to such studies. The Maryland Department of Assessments and Taxation uses the fundamental ratio statistical measures of the Standard and has adopted IAAO’s Assessment Ratio Performance Standard as the criteria to judge the performance of Maryland revaluations.

The IAAO Ratio Performance Standards are:

Ratio Study Performance Standards

Source: Standard on Ratio Studies; International Association of Assessing Officers; Chicago, Illinois; July 1999; pg 34.

Ratio studies may be performed for various reasons including appraisal accuracy and assessment equity studies, to judge the need for management of a reappraisal, to identify problems with appraisal procedures, to assist in market analysis, and to adjust appraised values. Many ratio study design issues must be considered depending on the purpose of the ratio study.

This study considers unadjusted sales price data six months prior to and six months after the date of finality (date of valuation, January 1st) for which assessments have become effective so that an unbiased estimate of assessment performance can be obtained. Only sales that are arms-length transactions between willing and informed buyers and sellers are included in this study. Maryland’s ratio performance is good and conforms to the IAAO Standard.

While several measures of central tendency are calculated (average, median, and weighted ratios), the median is less affected by extreme ratios. Therefore, the IAAO observes in its Standard that the median is generally the preferred measure of central tendency for monitoring appraisal performance. For this reason, median ratios are used in this study to measure compliance with IAAO standards.

The 2004 Ratio Study finds that, for residential properties, 4 of the 24 counties met all applicable measured IAAO standards, 17 met all but one, and 3 missed on two standards. All counties met applicable measured IAAO standards for coefficient of dispersion which indicates statewide uniformity in assessments.

As a proxy for time adjustments, this report uses sales from six months before the date of finality to six months after the date of finality. Under normal circumstances, with steadily changing property values, these sales will balance. In unusual circumstances, when property values are rapidly changing, this will affect the ratio statistics. The last two years have been unusual. The proxy for time adjustments does not work in a market where residential values have increased by double-digit percentages for each of the past two years. This causes the measures of central tendency, mean ratio, median ratio, and weighted ratio to trend downward.

Maryland’s local jurisdictions experienced very competitive real estate markets in 2003 and 2004. The desire to live in these jurisdictions, combined with interest rates favorable to buyers, has fostered continued large increases in sales prices. The largest growth continues in the Washington, D.C. and Baltimore City suburbs. These jurisdictions include Anne Arundel, Baltimore, Calvert, Carroll, Charles, Frederick, Harford, Howard, and Prince George’s Counties. Waterfront areas adjacent to the Chesapeake Bay lure buyers to Anne Arundel, Baltimore, Dorchester, Queen Anne’s, and Somerset Counties. Cecil and St. Mary’s Counties, once considered primarily rural in nature, have become a part of the suburbs. Many people are choosing to live in Caroline and Queen Anne’s Counties on the Eastern Shore and commute daily to the Western Shore.

Quality of life combines with house and lot size to draw more buyers to the outer suburbs and the Central Eastern Shore. Higher property values in the inner suburbs are creating booming real estate markets in the outer fringe of the metropolitan area. The lowest interest rates in 40 years, a low inventory of homes for sale, and multiple buyers competing for houses on the market have fueled an upward spiral in housing prices.

Washington County, traditionally thought of as Western Maryland, is beginning to experience growth in residential values due to proximity to Frederick and Montgomery Counties. Property values in Group 1 in northern Worcester County have also escalated. Maryland’s ocean resort area continues to attract buyers from throughout the mid-Atlantic region.

In addition to soaring property values in the Baltimore/Washington region, waterfront property values have continued their upward trend due to demand combined with scarcity. Since waterfront properties are generally higher-priced than inland property, this has affected the price-related differential in jurisdictions that revalued waterfront property during this assessment cycle: Dorchester and Somerset Counties. Baltimore City revalued properties in the northern portion of the City. The more expensive homes in that part of the City are more likely to be older, unique properties. Many attributes which may increase selling prices are not readily apparent during an exterior physical inspection and may not be captured in the assessment roles, affecting the price related differential. In counties with a smaller number of higher priced properties under reassessment such as Caroline and Garrett Counties, any factors which may not be captured for the higher priced properties inflates the price related differential.

All counties met the IAAO standard for coefficient of dispersion, indicating an overall uniformity of assessments. Although the measures of central tendency in some counties are low, the measures of variance are also low. This indicates uniform assessment levels

Statewide commercial properties have shared in the recent increase in real estate values. Commercial properties are generally less similar than residential properties. Many commercial properties are income producing and are valued using the income approach to value. Most commercial uses are cyclical in nature. Various segments of the commercial real estate market may be ascending in value as a class, while others may be declining in market popularity. Because of the uniqueness of commercial and industrial properties, measures of central tendency tend to vary more widely than with residential properties.

The number of commercial properties is small compared to the number of residential properties. In several jurisdictions, the number of commercial properties which have sold is so small that the statistical measures are prone to bias. Allegany, Caroline, Garrett Kent, St. Mary’s, and Somerset Counties all had fewer than 10 arms-length commercial transfer for Group1. In those jurisdictions, individual statistical measures would be unreliable due to sample size.

The commercial market in Harford County is advancing due to their very convenient location to many of the urban centers in the Northeast and Middle Atlantic States. Proximity to the Interstate Highway System has lured new businesses to the Interstate 95 corridor.

Commercial properties in northern and southern Prince George’s County and northeastern Anne Arundel County have experienced increased development and demand. Residential development in Calvert and Queen Anne’s Counties has increased demand for commercial property.

One impediment to commercial valuation is the increased use of the transferring of the controlling interest of the entity which controls the real estate instead of the use of deed recordation. This decreases the pool commercial sales available during valuation. It also may create a downward trend in assessed values due to lack of market data.

When fewer benchmark commercial properties which record deeds at the time of sale and disclose the sale price of the property, the quality of market data for commercial property is degraded. This phenomena is occurring statewide.

SECTION IV – STATEWIDE COMPARISON OF DEPARTMENT’S VALUES TO SALE PRICE

Quality is the degree of excellence of a product or service; the extent to which it measures up to certain standards. In this case, a measure of quality is the ratio study measuring whether the assessor appraised properties uniformly at market value. The ratio study conducted in this report is based upon sales data occurring, for the most part, after the time period of sales used by the assessor in the group of properties being reassessed.

Assuming the assessor applied the mass appraisal model uniformly to all properties, this ratio study should show uniformity of assessment. This ratio study is a cross check by Department management to assure quality of the mass appraisal work product. The ratio statistics for each county in Table IV was conducted on 36,530 improved residential property sales from July 1, 2003 to June 30, 2004 and compares the Department’s valuations to sale prices.

The frequency distribution in Table VI and statistics following present a statewide ratio analysis of improved residential property sales from July 1, 2003 to June 30, 2004, comparing the Department’s values to sales prices. The measures of central tendency indicate that properties are valued at approximately 88% of sale price and that on average all other properties have very similar ratios as indicated by the 11.25 Coefficient of Dispersion. Uniformity is also indicated by the number of ratios in the frequency close to the 90% level. Additionally, higher valued properties are assessed at a similar level to lower valued properties as indicated by a Price Related Differential statistic of 1.01.

The analysis from Table VI and the following descriptive statistics indicates that values determined by assessors for the most recent triennial Group 1 valuation attained a uniform and appropriate level of value. At the time of valuation, the assessments were close to the sale price. However, in an escalating market, assessments will quickly be surpassed by sale price. In summary, the data show that properties throughout the State are assessed uniformly.