Reader Challenges Skinner on Property Tax Expertise

Someone calling himself “Robert Williams” wrote the following in a comment, which I think deserves a reply:

When someone like Cal Skinner starts rambling about property taxes and assessments, I like to check his “facts” with my Township Assessor.

My Assessor gave me the Intnational Association of Assessing Officers (IAAO) standards for C.O.D.s, and they vary by property type and sample size.

For residential property with homes on it, a C.O.D. of 20% is only acceptable for “Rural or small jurisdictions / older properties / depressed market areas.”

So, apparently it IS on the spectrum.

There are whole different C.O.D. standards for income-producing property, vacant land, and so on.

There’s also something called the P.R.D (price related differential) that tells if you if cheap homes are overassessed and mansions are being underassessed.

So I guess that, again, Cal is an “expert” in something he knows little to nothing about.

Big surprise…

My reply:

First let me point out that Mr. “Williams” does not reveal the township in which he resides.

So, his source, his local assessor, remains anonymous.

In the late 1960’s I took and passed the International Association of Assessing Officers (IAAO) course that allowed me to be designated a “Certified Illinois Assessing Officer.”

While I was McHenry County Treasurer in the late 1960’s, I handled appeals for overassessed taxpayers that went all the way to the Illinois Property Tax Appeal Board.

All won their appeals.

To my delight, I was able to hand out a refund check for $500 to a Crystal Lake Coventry Subdivision resident.

The next year, the SPTAB required those handling appeals to be attorneys.

If you are satisfied with a margin of error of 20% where you live, that’s fine.

In my opinion, it is way too high.

Only Grafton, Algonquin, Nunda, Dorr Burton and Richmond Townships have margins of error under 20%.  

At a 33 1/3% assessment level, A Coefficient of Dispersion of 20% means that half of the properties are assessed between 26.7% and 39.9%.

The rest are either assessed higher or lower.

In 1973, I passed the bill that required members of Boards of Review to pass a test for the first time. I then took the test and passed with a score of about 90%.

Spreading the farm assessment in effect for DuPage County to all counties was my bill.

I served on on the Conference Committee that lowered the assessment level to one-third of market value, making a concession to allow a three-year average that I very much regret.

I proved that the State Aid to Education formula was discriminating against overassessed counties like McHenry.

And I was McHenry County Treasurer for four years (1966-70).

If that doesn’t qualify me to “rambl[e] about property taxes and assessments,” I can’t imagine what would.


Reader Challenges Skinner on Property Tax Expertise — 64 Comments

  1. Rambling = free speech.

    Sadly this site and generally speaking suppression of free speech is way to common even it this country.

    Of course not telling the whole story sucks too.

  2. Personal peccadillo.

    Mine is that I cannot endure morons or nitwits that misuse the word to and don’t understand too.

    (Nob used it twice and only scored 50% on correct usage.

    One shows action.

    One means more.

    The one that mans more has an extra “o”!

    How difficult is that to understand?)

    Cal’s panties are in a knot righteously.

    He is as close to an expert as you can get in these here parts on taxes.

    Robert Williams just garnered himself a hash mark for the moron club.

  3. Come on Cindy by now you should know English is like a secondary language to most of us.
    Chill a tad too.

  4. As an unbiased perspective, here’s what the Illinois Property Tax Code says about sales ratio median and COD accuracy (35 ILCS 200/4-20):

    (1) the median level of assessment must be no more than 35 1/3% and no less than 31 1/3% of fair cash value of property in his or her assessment jurisdiction; and

    (2) the coefficient of dispersion must not be greater than 15%.”

  5. Below is the excerpt from what I believe was referenced.
    Sources: Guidance on International Mass Appraisal and Related Tax Policy.

    “The following standards are recommended for the COD:

    • Single-family homes and condominiums: CODs of 5 to 10 for newer or fairly similar residences and 5 to 15 for older or more heterogeneous areas

    • Income-producing properties: CODs of 5 to 15 in larger, urban areas and 5 to 20 in other areas

    • Vacant land: CODs of 5 to 15 in very large areas with active markets, 5 to 20 in large to mid-size areas with slower development, or 5 to 25 in rural or seasonal recreation areas

    • Rural residential, seasonal, and manufactured homes: CODs of 5 to 20

    • Rural vacant land with little development: CODs of 5 to 30.”

  6. Sure, accurate assessments are part of the property tax structure, but that is not the root of the problem.

    The root of the problem is wasteful spending on education: social workers, esl teachers, counselors, and administrators coming out the yahzoo to collect every tax dollar possible from “grants”.

    On top of the waste are State Pension and Prevailing Wage laws plus all the lawsuits and absurd workman’s comp costs.

  7. Mr. Z., with so much of the assessment process formulaic, what is the justification for so many human assessors?

  8. IF a township is correctly classified as “Urban,” then Cal is indeed right that its COD should be under 15%. Anything higher is unacceptable.

    However, looking at the listed data, someone of Cal’s stated expertise should know that if an entire Township has only 49 sales over a THREE-YEAR PERIOD (like Burton), it is either a “Rural” area incorrectly classified as “Urban” OR it is an “Urban” area with “depressed market conditions.” Either status makes 20% an acceptable COD. Marengo with 68 sales over THREE YEARS, and Richmond with 67, also suggest the above. “All others” means the remaining FIVE townships in McHenry County, with a measly total of 123 sales OVER THREE YEARS between all five, again suggesting that “All others” are also incorrectly classified (and we cannot see the township-by-township distribution of data). There’s no excuse for Chemung Township’s COD of 29.96, even assuming that it, too, fits the above.

    My Assessor also tells me that the PRD (price-related-differential) is missing from the listed data. This is a measure that tells you whether or not lower-market-value homes are being over-assessed and whether higher-market-value homes are being under-assessed.

    And no, Cal, I am not revealing my Assessor’s name. My Assessor hasn’t made any death threats, unlike others on your blog, and also doesn’t deserve to be exposed to the vindictive nature of the McHenry County powers-that-be.

    And as a historic side-note: why were you, as McHenry County Treasurer in the 1960’s, representing appellants in property tax assessment appeals?

    Was that part of the Treasurer’s duties back then, some kind of inherent official conflict of interest that has since been eliminated?

  9. Oh, excuse me, quick edit: make that “alleged death threats…”

  10. Including the far right hand column of price-related-differential information would have made the chart numbers a lot smaller.

    Stay tuned.

    Still interested in the anonymous township.

    My representation was not in my official capacity, but everyone who appealed to the State Property Tax Appeal Board won.

  11. Cal, can you explain why/ (if) so many different human assessors are necessary when the assessment value as described above can be derived given these computerized algorithms?

    Has there been a “computer vs. human ” comparison of accuracy” performed?

  12. Asking why humans are needed to properly apply the appropriate processes is like asking if the F-22 Raptors are on the ramp and they’re so smart why do we need pilots.

    Give a power floor sander to a D-I-Yer and the odds favor an irreparably damaged floor.

    Give the same machine to a skilled professional and one gets a floor finely polished and ready for sealing.

    The proper tools alone don’t achieve the desired end.

    The proper tools in the hands of properly educated humans do.

    To the best of my knowledge, judgement is one of the few traits where humans still outshine computers.

    Besides someone has to develop the assessments that feed the DoR’s statistics machine.

  13. The designation of neighborhood, township, etc. being appraised has nothing to do with the number of sales during some arbitrary time period.

    It has to do with the development status of the location.

    There are three designations: urban, suburban and rural.

    Per the The Dictionary of Real Estate Appraisal each is defined below.

    Urban – describes a mature neighborhood with a concentration of population typically found within city limits or a neighborhood commonly identified with a city.

    Suburban – describes a neighborhood that contains complementary properties with less concentrated population than is typically found in an urban neighborhood.

    Rural – pertaining to the country as opposed to urban or suburban; land under an agricultural use; areas that exhibit relatively slow growth with less than 25% development.

    In addition, the standard Fannie Mae appraisal form (1004) provides the following check boxes:

    Urban: more than 75% built-up.

    Suburban: 25% to 75% built-up.

    Rural: less than 25% built-up.

  14. What is the number of properties typically covered by one assessor?

    Each township has an assessor, but vary in number of properties .

    Also, 100% of assessment appeals ask for lowered (not raised) assessment, but 100% of error are not deemed to be over-assessments.

    What mechanisms are applied to correct under-assessment errors, which damage all other taxpayers?

  15. For the system to work, tax districts need to appeal underassessments.

    Back in the 1970’s Algonquin Township Assessor Forrest B. Hare arrayed the margins of error of all assessment jurisdictions in Illinois.

    Although there were exceptions, the worst assessments tended to be in townships with fewer than 5,000 residents.

    My guess is that with more people than that a full-timer could be hired, which resulted in a more qualified assessor.

    Based on that research, I sponsored the Multi-Township Assessor bill.

    It allowed township boards to consolidate the function through the election of a Multi-Township Assessor.

  16. Susan is correct: assessor’s are township-based not parcel-based although not every township has a full-time assessor.

    Some are part-time (a very poor ROI for taxpayers) while some have contracted assessors.

    And, as Cal states, the tax code allows multi-township assessors where one assessor assesses more than one township.

    Regarding correcting under-assessments, that generally can only be done on a township or neighborhood basis usually in a general assessment year unless things are really out of whack.

    Every time we’ve asked for an increase (versus the appellant’s request for a decrease) at a Board of Review hearing we’re laughed out of the room.

    None of the Deputies can remember ever seeing an assessment increased at the Board of Review.

    Vegas would go broke if people knew they could gamble with no downside.

    Just like appealing: the worst that can happen is a no-change.

    The truly damaging assessment reductions are large commercial pins.

    In addition to shifting their significant tax burden to everyone else, they’re so large that enough of them can even influence a township’s equalized assessed value (EAV).

    Since the tax rate and EAV are inversely proportional, as their (and the township’s) assessments decrease, the tax rate increased for everyone.

    It’s a battle we fight, and sadly lose, every year.

  17. What about oblique solutions?

    If you agree that the problem can’t be solved within current political system, is there a way to introduce a program which creates incentive/disincentive by way of independent competitive forces?

    Such as: Once a year,, every building* could be (sealed) bid upon at amounts (at least 20%**?) higher than assessed value (not including exemptions).

    The owner could retain ownership by agreeing to the higher assessed value (of a willing buyer)but getting to apply exemptions to lower the assessment again as always.

    *property taxes rise as a function of social service provision obligation. buildings, rather than land, demand schools police fire protection etc.

    **in order to protect homeowners from greenmail (costs of moving may exceed differential), bids should be sufficiently higher than current assessment.

  18. Great idea deserving consideration.

    However, tying ownership (deed) to property tax is a last resort, e.g., tax sales.

    My suggestion is an Ockham’s razor solution: just have everyone obey the tax code and the Board of Review rules as they’re written.

    Both are very clear.

    If there were penalties for assessors, Chief County Assessment Officers and Boards of Review who didn’t obey them, there’s half the solution.

    Adding disincentives like penalties for extraordinary assessment errors is the other half.

    If assessors and CCAO’s are rewarded for good performance (35 ILCS 200/4-20 and 35 ILCS 200/3-40), why not penalize those who fail even the most basic standards of accuracy (much less the blatant shenanigans uncovered in the Tribune article)?

    The argument is already ringing in my ears:

    “Valuations are opinions. How can you penalize someone for their opinion?”

    My response would be:

    “Yes, valuations are opinions. However, they must be based on statutes, rules and accurate data.

    “When the first two are heeded without variance, and the the third are credible, the variance of opinions would be minimal.

    “Therefore, don’t measure one’s opinion; measure one’s adherence to the statutes, rules and data credibility.”

  19. “How can you penalize someone for their opinion?”

    1. Tell that to a medical doctor.

    2. You penalize all taxpayers for opinions, when ‘opinions’ err to the low side.

    What about, an assessor (district) who can be shown to have X-number/X-percentage under-assessed properties without back-up rationale (by some equitable metric) is perfunctorily defaulted to authority of County Assessor?

  20. The “authority of County Assessor” is where the previously-mentioned massive assessment reductions for large commercial pins are occurring.

    That’s one good reason not to consolidate assessments at the county level: only one set of eyes looking at the numbers.

    A second good reason is during one of its more lucid moments, the legislature required township assessors to live in their township.

    That provides likely the greatest benefit for property owners: geographic competency.

    The Uniform Standards of Professional Appraisal Practice (USPAP) require this of all appraisers via its Competency Rule.

    The current 2016-17 version of USPAP currently in force states in lines 372-374:

    “In an assignment where geographic competency is necessary, an appraiser who is not familiar with the relevant market characteristics must acquire an understanding necessary to produce credible assignment results for the specific property type and market involved.”

    In Grafton, we have our hands full correctly understanding the roughly 80 distinct neighborhoods we have. The nightmare of trying to competently understand an entire county (~150,000 parcels in McHenry County) is incomprehensible!

  21. It is called bureaucratic paper shuffling; and it is the death of all “civilized” society.

  22. Correct me, but voters will tend to vote in favor of assessor who assesses on the low side of ‘accurate’?

    Built-in skew?

    Would you as a confident assessor, submit to a ‘contest’ with ‘Watson’, to see whether human Assessors can beat a Computer (like Ken Jennings on Jeopardy or Watson vs. Kasparov chess match)?

    Remember, taxpayers shouldn’t measure results relative to ‘Perfect Outcomes’, taxpayers should measure
    imperfect outcomes generated by a computer impervious to greed, fear, and political/personal pressure vs. CURRENT IMPERFECT OUTCOMES of human assessors subject to greed, fear, and political/personal pressure.

  23. Can an Assessor quantify the nominal amount of tax revenue lost/gained due to:

    ‘unfairly LOW assessed properties’


    ‘unfairly HIGH assessed properties’

    based upon COD?

  24. Voting in favor of low assessments is self-defeating; just look at what happened in Cook County as depicted in the Tribune articles when properties were under-assessed by 40%.

    Besides, ultimately it’s a zero-sum game:
    Levy/township EAV = tax rate.
    Tax rate * net assessment = property tax.
    Lower assessments times a higher tax rate (based on the lower EAV) is the same as accurate assessments and an accurate tax rate (based on the real EAV).

    Our office would be glad to take on Watson!
    Reasons being:
    – we use several of his very low CPU-horsepower siblings and regression anlysis to create our assessments
    – we have the human judgement Watson doesn’t which is why appraisers beat AVMs day-in, day-out.

    I can/will state with absolute certainty each of the “human assessors” in Grafton Township are completely immune to “greed, fear, and political/personal pressure.” Our mantra is “Know what we’re doing. Never cut a deal, never will.”

    Anyone can “…quantify the nominal amount of tax revenue lost/gained…” and one doesn’t even need the CoD.
    From a township, county or even state perspective:
    (Correct tax rate * correct assessments) – (Biased tax rate * incorrect assessments).

    The levy stays the same regardless and the tax rate and EAV float with the assessments.
    Two equations, three variables = a properly (versus an under- or over-) specified, and thereby solvable, equation.

  25. Al Zielinksi: Your an Assessor, I’m getting my info secondhand from my Assessor, BUT would you expect a “75% built-up” Township of 36 square miles to average 22 sales a year?

    Susan: I’m the owner of an IT company.

    If people bought homes using only computer algorithms, THEN you could make all real estate appraisal AND tax assessment computer-based.

    Zillow is facing a firestorm right not because of their reliance on computer models results in grossly inaccurate real estate value estimates.

    Because the home market is not like buying soy bean futures – human judgment, psychology, and tastes are just as big a factor as location, location, location.

    Nobody ever paid more for soy beans because their wife liked the color scheme…

    Cal: Being County Treasurer and repping tax assessment appeals at the County and the State takes more…something…on your part than I have.

    What if you had lost?

    “The County Treasurer botched and lost our tax appeal for us” might have been the least of the potential accusations of liability from the taxpayer…

  26. I read the railroad assessment appeal cases which said that railroad property should be assessed at the same percentage of market value as other county property.

    At the time the Supervisor of Assessments, Stanley Cornue, had issued instructions to assess property at 60% of market value.

    In reality, only the newly-purchased homes were set at that level.

    Using two different methodologies to compare assessments to sales prices, I found the average assessment in McHenry County was 42.3%, if memory serves me correctly.

    To lower the 60 percenters to the county media was pretty much a no brainer for the State Property Tax Appeal Board.

    The experience also convinced me that members of Boards of Review needed to prove they knew someting about the law before being allowed to pass judgment.

    Indeed, after I was out of office a property owner asked me for company at a Board of Review hearing.

    As the former State’s Attorney Dick Cross saw me enter the room, he said something to the effect that the Board better get ready to lose.

    The homeowner won, but I can’t remember whether it was there or at the SPTAB.

  27. As previously stated, the “built-up” designation of a neighborhood has nothing to do with the number of sales in any time period.

    It relates solely to the extent of development and is usually bracketed:

    >75% = urban
    25-75% = suburban
    <25% = rural.

    When parameters are published by entities like Fannie Mae, Freddie Mac, etc., it's best to stay with those industry norms versus substituting personal opinions.

    Regarding the comment on Zillow (and other AVMs), I completely agree.

    That comment also proves (at least) one point of this article: if AVMs use public data as their input, those data better be accurate!

  28. What query I failed to articulate properly was, in the process of :

    “Anyone can “…quantify the nominal amount of tax revenue lost/gained…” and one doesn’t even need the CoD.

    From a township, county or even state perspective:

    (Correct tax rate * correct assessments) – (Biased tax rate * incorrect assessments).”

    —-are inaccurate low assessments of high end homes being blended with disproportionately inaccurate high assessment of low end homes in order to smooth the curve?

    For an anecdotal example,the arguably $4 million low assessment of Pritzker’s mansion-adjacent mansion would be balanced out by 200 ‘average’ homes whose assessments were $20,000 too high, but no lawyer would bother to take their tax appeal case for 1/2 of 2.3% of $20,000 contingency fee?

  29. RW, what is the relative comparison of cost benefit from starting at a new baseline, given that all those who perceive their assessment has been unfairly high have appealed, and all those who feel their assessment has-been fair or low have declined to appeal, then going forward based upon baseline vs. New sales data?

    Rather that compare to some unobtainable ideal, I am suggesting we compare a less imperfect system to the grossly imperfect system we suffer under presently (Grafton Township excepted, I guess).

  30. Susan if I correctly understand your “blending” question, no.

    My understanding is all assessment (at least within a township) are taken as discreet data points.

    The only “blending” that comes into play is the township (and possibly state) equalizer(s) because they are calculated on a township (and state) level.

  31. Let me re-frame the question.

    1.The current system–human assessors–allows some amount of assessment inaccuracy which results in friends of assessor rewarded, enemies of assessor punished severely, and those to whom assessor is indifferent punished moderately.

    2.An alternative system–computer assessors–allows some amount of assessment inaccuracy which results in non-emotionally-generated rewards and punishments, presumably equally and randomly distributed.

    1.Current system, human, has inaccuracy challenged when it is too high and never when too low.
    Tax rate is driven ever higher as a result (numerator the same or rising, denominator only decreasing).

    2. Alternative system, computer, would have fewer challenges because the bases for challenges are data which comprise the assessment algorithm.
    Tax rate may be closer to fair value as result?

    Question is this:

    Would you please describe to the public once and for all an objective, data driven method for value determination so that we can all get the same deal we perceive the insiders are getting?

    That is, if bricking up a chimney or unplugging a toilet can result in a gigantic reduction in assessment such as Pritzker received, we all want to know.

    If there is a formula that is a function of age of home, price per square foot, lot size, etc….we all want to know.

    (Not everyone is able to get the attention of busy, politically connected, expensive assessment appeals attorneys: the smaller the home value the less likely a homeowner will get representation.)

    If an Assessor CAN’T describe such a formula, and it truly is such a subjective and infinitely variable problem with infinite expense bounds that justify the human error/human cost, then would you recommend citizens get busy working on a COMPLETELY alternative tax model which has a chance better than a fart in a windstorm of attributing fair share tax burdens to all taxpayers?

  32. Forget about statistics; it’s not about a formula.

    It’s about common sense.

    And it will come at a “cost” (pennies on the tax dollar) through higher salaries for assessor staff at the township and county levels.

    1. Hold everyone’s feet to the fire.

    Everyone = township assessors, Chief County Assessment Officers and Boards of Review.

    Fire = if adequate evidence is presented to the State’s Attorney, charges of misfeasance are filed and prosecuted.

    Misfeasance = the defendant owed a duty of care toward the plaintiff(s), the defendant breached that duty of care by improperly performing a legal act and the improper performance resulted in harm to the plaintiff(s).

    If it’s REALLY bad (Cook County if the Tribune is correct), escalate to malfeasance.

    2. Require minimum education requirements for township assessors, Chief County Assessment Officers and Boards of Review.

    Right now = breathing and a CIAO designation.

    Minimum = high school education with some college level math and statistics courses.

    Better = college degree required with multiple college level math and statistics courses.

    Best = certified real estate appraiser by the state. This addresses the above education requirements.

    Technology is here to stay.

    The days of using the “Modified Cost Approach” to accurately assess primarily residential properties are long gone.

    Education, technical knowledge and applied statistics are required for accurate assessments today.

    It’s not an easy mathematical solution but you asked for clarity. This is as honest and open as I get.

  33. I certainly appreciate your thoughtful responses.

    This is teaching me so much.

    Thank you very much for taking the time to make this information available to me and every other citizen looking for specific solutions.

    Your ‘enforce the law’ solution sounds good until you try to find WHO has ‘duty’ to enforce the law, and can execution of that duty be demanded under the law or is it (politically) discretionary.

    Your education solution does nothing to address the impacts of greed and fear on human nature.

    I’m suggesting, maybe a few bad apples spoiled the barrel with patulin.

    I believe now we have gone too far in Illinois and cannot correct with humans involved.

    If we (Illinois Counties) were to incorporate a computer assessment system, with full appeal rights as exist today, how much worse could it get than it is now? Would the magnitude of error be the same but the demographic of the ‘victims’ shifted?

    Can we quantify that scenario to understand how we might defend ourselves against a tidal wave of Illinois political power driving taxation higher through the path of least resistance (homeowners paying property taxes)?

  34. Duty is clearly presented in the property tax code.

    Township assessor’s must sign an affidavit attesting (s)he did her/his job (35 ILCS 200/9-230):
    “…and that the assessed value set down in the proper column opposite the descriptions of property is a just and equal assessment of the property according to law.”

    Ditto for Chief County Assessment Officers: 35 ILCS 200/9-245.
    Boards of Review are covered in 35 ILCS 200/16-85.

    The issue is enforcement via penalties that sting.
    I’m no lawyer but it seems perjury becomes involved in addition to misfeasance when one signs a sworn affidavit.

    Greed and fear should be distilled out during the election process.
    Education must be complemented with character.

    Computers are always operated by humans; just look (again) at Cook County: that fiasco was accomplished with a few lines of computer code.

  35. I meant that no office of Illinois has enforceable duty to investigate, prosecute, and obtain restitution from bad actors.

    Poorly coded programs will produce objectively poor results which can be challenged by ‘victims’.

    Code can be examined and changed on a broad scale so that everyone is in the same boat–good boat or bad boat.

    What can’t happen with code, without leaving a clear footprint to actionable fraud, is for a single connected developer to obtain extraordinary individual treatment obtained by law firms fast tracked to an Assessor to whom they have personally donated hundreds of thousand$ in campaign contributions.

  36. Again, I’m no lawyer but I thought that’s what States Attorneys do: bring civil misfeasance cases.

    Affidavits are much more easily understood by the courts and juries than computer code.

    If three entities sign separate affidavits clearly stating what they submitted is a
    “…just and equal assessment of the property according to law.”
    shouldn’t that be good and straightforward enough to prosecute them if the numbers are provably wonky?

    If not, let’s pack-up our tents and move to some place where it is.
    If nobody is held legally accountable for something as significant as the ad valorem valuation of real property, game over.

    To see what happens on the other side of the fence (appraisers) peruse the Illinois Department of Financial and Professional Regulation’s discipline section and look under “Appraisal.”
    That’ll put the fear of God in anyone with half a brain.

    If that level of enforcement is available to the public for fee appraisers, why not mass appraisals?

  37. Susan: I own an IT company and I am a programmer.

    The only way for a computer program to do what you want it to do is for someone to code human buyers’ psychology, wants, and motivations into an objective and reliably predictive computer model.

    For instance, charging a uniform rate per square foot for all two story homes.

    All two story homes, be they 130 years old and falling apart or brand-new custom mansions, get assessed on identical square-foot costs.

    Is that fair?

    So now our program has to adjust accurately for differences in age and condition.

    And what if the 130 year old poor-condition two-story is located on land in a prime downtown Commercial center and the mansion is located next near an industrial park that includes a meat processing plant?

    Now the program also needs to accurately adjust market values for differences in location and highest and best use.

    What if the owners of the poor-condition two story decide to invest in massive upgrades for the home, while the mansion begins suffering from chronic roof leaks, rotting window frames, and flooding in their superbly finished basement due to deferred maintenance? Now the program has to accurately adjust for maintenance, too.

    And we’ve only started to consider all the variables in the real estate market.

    If I could develop software that could deal with such a multitude of variables and correctly predict market value with no human interaction, I wouldn’t be wasting it on real estate appraisal (or mass appraisal) – I’d be coding it around the financial markets and making REAL money…

    Al: you and I and my Assessor all know that if an “Urban” 36-square mile township, one that is supposedly “75% built-up,” has only 22 sales in a one-year period, SOMEBODY had better take a look at the underlying data.

    Because either Robert Ross (or the State Department of Revenue) have incorrectly classified some these townships as “Urban,” OR they are suffering from a blighted, depressed market.

    Either condition would justify a COD of 20% (the original point of the conversation) under IAAO standards.

    And Al?

    When it comes to reassuring the public that tax assessments are accurate and uniform, and establishing draconian legal punishments for any Assessors who seem to violate that, YOU might want to consider whether YOU really want to start pulling at that thread…

  38. I’ll stick with the widely-accepted (and previously conveyed) definitions that “Location” has Urban, Suburban and Rural check boxes on the standard appraisal form.

    I’ll also restate none of those options have anything whatsoever to do with the number of sales during some period because they all relate to “Location.”

    There can be one sale in downtown Chicago and it would still be correctly classified as “Urban” because that definition refers to location density not sales.


    I set neither the determination to prosecute or the penalties if convicted.

    I merely pointed-out the mechanisms already in place should anyone decide to use them.

    Given the Department of Revenue’s analysis of Grafton’s performance, I’ll gladly be first in line.

  39. RW you are describing a finding the lion in the jungle problem with the assumption that nobody knows what a lion looks like.

    The starting point is not a set of complete unknowns with infinite variables.
    factors such as age, for example, can be quantified to some degree in a standard manner. Every year, each building gets one year older.

    Rehab work requires permits, and I assume assessors are notified now when that happens.

    Let’s compare the imperfect system we have now to the proposed imperfect but objective system proposed.

    Each are subject to judicial review if outcomes are perceived as unfair (unfairly high.

    I know of no mechanism today whereby citizens may initiate protest on perceived under assessment of others properties).

    But only human assessors are allowed to repeat the same ‘bad judgment’, because a program would be updated toward improvement, evenly applied to all similarly situated properties.

  40. The Supervisor of Assessments advise me:

    “Tax districts may file an under-assessment appeal on a Board of Review decision if the assessment reduction request is $100,000 or more. We have to notify tax districts on all complaints requesting the $100,000 or more reduction. Any tax district can then have standing on the appeal if they file as an intervenor. The tax district could also file an under-assessment with the Property Tax Appeal Board.”

    An individual taxpayer does not have that authority.

  41. Susan: Re-read what I wrote.

    Then take a software programming course centered around predictive analytics.

    And then ask yourself why every bank and mortgage company in this country has not entirely dropped the use of human appraisers in favor of free algorithm-based tools like Zillow, Redfin, etc.

    Al: Do you REALLY want to play semantics on this one?


    what “penalty” might you recommend if an Assessor-elect (who also had a private appraisal practice) offered to reduce the assessment of any taxpayer IN HIS OWN JURISDICTION who paid him for a fee appraisal?

    That is, of course, ONLY in the UNLIKELY event that the Assessor-elect with the private appraisal practice happened to arrive at a lower value than the taxpayer’s assessment?

    Maybe it wouldn’t merit a penalty at all, but instead deserve re-election?

    That’s Illinois for you…

  42. USPAP compliance prevents appraising to reach a predetermined value i it’s stated as such in the mandatory Certifications section of all appraisal reports.

    Therefore, your scenario would never happen with an ethical appraiser (or any appraiser who wanted to keep his/her license and avoid penalties from the IDPR).

    That’s why ALL fee appraisals I perform are USPAP-compliant.

    It’s also why I set and achieved the goal of USPAP-compliance for our office by the end of my first term.

    As I’ve stated throughout this thread, and will always continue to state, people deserve honest and accurate valuations.

    Regarding your comment to Susan, Freddie Mac recently removed the requirement for appraisals in certain situations and using their AVM as a replacement.

    Fannie Mae has been using Collateral Underwriter (a form of regression analysis) for years as a check on appraisals they receive.

    The VA uses a comparable process from CoreLogic.

    Two main issues with Zillow et all is

    (1) they use public data (GIGO) and

    (2) don’t complement it with skilled human judgement.

  43. RW I assumed that you knew there are programs already developed, in use ( or said to be uses but deliberately not in use in the case of Chicago’s Chief Assessor).

    My original question was, why does each township need a human assessor ( with variable numbers of property to assess per township, and with the temptation to subconsciously reward friends and punish enemies and strangers by offloading on them the responsibility of paying for the discounts given to under-assessed properties)?

    At the first pass of an assessment, whether human or using the UC generated new Cook County model( automated ), homeowners have the right to appeal.

  44. 1. Property owners can always appeal regardless of whether the assessments were human or computer generated.

    1a. Boards of Review and the Property Tax Appeal Board should heed and adhere to 35 ILCS 200/16-80 and 16-185 respectively.

    These deal with people who appeal yearly clogging the system. They both state “…reduced assessment, subject to equalization, shall remain in effect for the remainder of the general assessment period as provided in unless the taxpayer, county assessor, or other interested party can show substantial cause why the reduced assessment should not remain in effect, or unless the decision of the board is reversed or modified upon review.”

    2. Human assessors are needed to:

    2a. oversee the computers and the competent people who run them and

    2b. apply human judgement of which computers are incapable.

    2c. Computers can’t drive out to properties to measure them and determine if assessment changes are required. The Property Tax Code requires viewing of ALL parcel at least every four years (35 ILCS 200/9-155):

    “…in each general assessment year in those districts, the assessor, in person or by deputy, shall actually view and determine as near as practicable the value of each property listed for taxation as of January 1 of that year…”

    3. Bias and favoritism should be dealt with using the statutes and performance metrics already in place.

    3a. Oversight and correction already exist; the will to implement and enforce them are lacking.

  45. Because assessors are tied to townships by the Property Tax Code, the number of PINS floats significantly based on the township’s population density.

    You can see the wide variance in the number of parcels and EAV by downloading and reviewing the Chief County Assessment Officer’s Annual Report at:

    Beyond the statutory issue, rather than the number of PINS, their diversity plays a much greater role.

    Agricultural parcels are assessed by the state based on soil productivity so those are off the table.

    Grafton is primarily (~95%) residential but even those residential parcels required almost 100 neighborhood designations to achieve homogeneity.

    Commercial and industrial parcels really up the ante because “cookie-cutter” commercial and industrial parcels are few, far between and almost non-existent.

    They’re also were the big bucks are.

    Being “off” on some of those (like just one Walgreen’s) is far worse than being “off” on a few residential properties.

    Perhaps rather than tying an assessor’s qualifications to the number of parcels, it could/should be tied to the township’s Equalized Assessed Value (EAV): the greater the township’s value (and likely diversity), the higher the required qualifications for the township’s assessor.

    Right now, the qualification is homogeneous: a CIAO designation.

  46. Susan: As much as it pains me to agree with Al Zielinski, here’s the answer to your question:

    Because HUMAN JUDGMENT is an essential part of the appraisal and assessment process, as it is of ANY work involving market-based data.

    I understand that you share the general public’s touching, near-religious faith in computers as accurate, “fair,” and objective.

    As a computer programmer, I can tell you that no commercial software can take the place of human experience, judgment, intuition, and ability to synthesize.

    We’re spending billions on artificial intelligence to achieve this goal, but guess what: NOBODY is spending those billions to create an AI to assess your property for tax purposes.

    No commercial software can “automatically” stratify properties by style, condition, age, size, construction, lot size, geographic location, and all other variables that go into real estate value and derive a “fair” uniform square-foot cost for every property UNLESS it has an experienced human appraiser or assessor to guide it.

    Without that human expertise, only the gods of the chipset know what value will be placed on your property.

    Here, let’s make it simple: Call a realtor and sign an agreement to list your home for whatever Zillow’s says it is worth.

    Then go on Zillow.

    Get its algorithm-based value of your home.

    By Zillow’s own recent admissions, its values for single-family residential tend to be 20-25% below actual market value.

    Now sell your home, take your losses, and rejoice that you’ve personally sacrificed time and treasure for your faith in the Church of Cyber-Appraisal.

    If Zillow has valued your home way too high, then let the realtor have a good laugh at you and insist that he list it at the inflated price anyway.

    And then reflect on the fact that you want a computer to set your property tax assessment for you.

    And as to your concern that human assessors are fallible and some might even be corrupt: you’re right.

    Exact same thing is true of judges, police, doctors, lawyers, airline pilots, schoolteachers, journalists, mechanics, military personnel, firefighters, paramedics, and (gasp!) elected officials. Want to replace them all with computer algorithms, too?

    Because I have a few dystopian science-fiction movies you may want to watch first…

  47. (And Susan, as an IT professional, your and the public’s religious faith in computer technology is my bread and butter, so keep it up; I know for a fact that computers’ “superiority” to human skill sets in certain key fields is total BS, but I’m more than happy to take your money)…

  48. RW it is disappointing that you react in such a defensive emotional manner.

    Your mischaracterization of me and the general public indicates your contempt for others.

    Also it indicates bigotry and paranoia toward religion in general, which seems irrelevant to the problem at issue.

    I think you are too far removed from me and others to make such a presumptuous generalization on either religion or ‘faith’ in technology.

    If this is a sample of your powers of correlation, I would distrust your algorithms.

    But maybe you didn’t mean it the way it sounded. You seem to be strongly supporting status quo and indicating that improvements through tech are impossible ?

    In this one case, tech can’t increase productivity–we need as many assessors as we did without available technology ?

    Where they can look at Athena rather than spend hours driving, for example?

    As to your example, houses sell all the time at prices well above assessment valuations.

    (Tax Assessment values are not Zestimates and are not treated as such by buyers. Also assessments lag in a normal non-rate capped market and are known to usually trend low).

    The point is trying to distribute tax burden fairly and equally .

    If all homes were an equal 20% below true fair market value, that would still be accomplished.

    As the Tribune analysis pointed out, higher priced homes tend to be more under valued.

    Again the problem :human assessor make unevenly distributed errors.your comment questioning assessor integrity underscores this is a problem.
    So we need a solution which impedes human powers to selectively favor friends and punish enemies.

    Using the software in existence and used by assessors today as a starting point, is there a way to reduce the number of assessor necessary ( to save public money and reduce opportunity for human greed and fear bdriving assessment values individually .? Is there a way to track outliers and force human assessors to be held to ethical standards cited by AZ?

  49. RW, on re-reading my response I can see it is…not as objective as I wish it had been.

    This Blog presents opportunities for people who seek to root out injustice.

    It also, very fairly, offers opposing points of view the opportunity to defend the status quo.

    Many times there are commenters who do not have a winning argument to defend (inequitable) status quo, and they find that diverting the topic through personal attacks or straw man arguments works just as well.

    I should not have assumed you are such a commenter.

    I hope you will bring your considerable problem-solving talents to the complex problems which burden McHenry County taxpayers, who pay around 4% of fair market value property tax rates.

    You wrote:

    ‘And as to your concern that human assessors are fallible and some might even be corrupt: you’re right.

    Exact same thing is true of judges, police, doctors, lawyers, airline pilots, schoolteachers, journalists, mechanics, military personnel, firefighters, paramedics, and (gasp!) elected officials.

    Want to replace them all with computer algorithms, too?

    Because I have a few dystopian science-fiction movies you may want to watch first…”

    In response, Yes.

    I WOULD like to REDUCE THE CAPACITY FOR HARM TO OTHER HUMANS many of the aforementioned professions possess.

    In my understanding, many of those professions now do employ a great deal of algorithmic protocol in triage and interventions.

    Let’s center our attention on those professionals who DO NOT RISK THEIR LIVES AND/OR ALL THEY OWN when engaged in their jobs:

    Judges, lawyers, schoolteachers, journalists and elected officials (including assessors) all have special exemptions from civil liability carved out under the law.

    In my experience,Enforcement of penalties on these professionals which ARE allowed under the law are not enforced on a pari passu basis with other professions. Do you disagree?

    Can we all agree that it would be beneficial to all humans if fair and equal justice were meted out?

    Can we agree that the judgement of merit of a solution should not be based upon ‘ideal vs. suggested solution’ but rather ‘status quo vs. suggested solution’?

    coming back to assessors, we have possible solutions:

    change assessment model

    enforce penalties on assessors who breach ethical application of rules
    establish bonus rewards system for assessors who are most correct, and rewards for seeking to emulate THEIR model.

    establish a reward system for citizens who may discover and reveal unethical under-

    assessments granted by human assessor using a non-uniform basis

    establish a truly objective (non-political) appeals system which can be accessed by any
    economic level of property owner on a practical basis.

    I’ll bet you can think of more, and better, possible solutions to add to the list.

  50. “change assessment model” YES

    Residential parcels should use the Sales Comparison approach.

    Requires assessors trained and skilled in regression analysis.

    Commercial/industrial parcels should use the Income Approach.

    Requires assessors trained and skilled in discounted cash flow analysis and property owners providing their business tax returns.

    The Modified Cost Approach is easy to implement but highly inaccurate.

    “enforce penalties on assessors who breach ethical application of rules” YES

    Also add incompetence.

    “establish bonus rewards system for assessors who are most correct”
    Already exists for township and county assessors: 35 ILCS 200/4-20 and 200/3-40.

    “establish a reward system for citizens who may discover and reveal unethical under-assessments granted by human assessor using a non-uniform basis”

    Ok by me as long as only credible claims were escalated to formal review.

    Otherwise, it’ll bury the assessor’s staff in bogus claims.

    “establish a truly objective (non-political) appeals system which can be accessed by any economic level of property owner on a practical basis.”

    Because Boards of Review are appointed by County Boards, good luck.

    I hope lots of people are reading this thread because it provides a wealth of positive dialog and education regarding the property tax and assessment system as well as well-considered changes.

  51. None. It floats by location. For example: Cook County = 10%, Wisconsin=100%.

  52. More precisely, Cook County residential is 10%?

    Cook County commercial / industrial is higher?

    Any idea of the history or reasoning why Illinois assess at 1/3rd of market value, instead of 100%?

  53. Thanks AZ for taking time to post constructive information.

    Any suggestions on how to accomplish forward progress given political constraints?

  54. Mark, I have always maintained a cynical assumption of ‘complicated by design’ to produce greatest profits to those hands through which the oceans of tax money flow and so strips of currentcy may be subtly diverted and directed one way or another. 1/3 makes an assessment seem very low at first glance to an unsophisticated homeowner. But maybe it was innocent, it is a way to make large numbers more manageable and understandable, if it removes a zero.
    Apparently there was a uniform under-assessment of ALL properties, and maybe that made it harder to appeal (at least on one grounds: fair market value)?
    If assessments were appealed, maybe that used to be considered a bad thing? Maybe it was politically detrimental at polls?
    But seeing as there was no penalty for poor performance,and as noted in Trib article posted on this blog, in 2010 the shift occurred to do away with across-the-board under-assessments and appeals burgeoned.
    This was due in part, the article says, to the real estate crash at the time, but: nominal taxes rose during a period of plummeting home values.
    This divergence (falling values/rising taxes) signals either levies rose (taxing bodies took advantage of global economic chaos to secure the salaries, guaranteed defined benefit plans with COLA requiring little or no contributions from public employees, guaranteed health insurance and dental and life insurance premiums paid on behalf of employees, OPEBs) or real estate values fell in a NON-UNIFORM fashion as dictated by assessors, or both.
    The Trib article points out that windfall profits were reaped by law firms engaged in the new big business of tax appeals/ Chief Assessor Berrios’ campaign funds received quite a bit of money in contributions from these very law firms.

  55. Ideally the median in chart in the blog post would be 33.33% (1/3 of market value).

    Which would indicate assessed value = market value.

    The formula for the median is assessed value / market value.

    What are some of the reasons the median values for all townships in the chart are less than 33.33%?

    Urban in the chart means only parcels categorized as urban are included in the study; not that the entire township is categorized by the Illinois Department of Revenue (IDOR) as urban.

    The sales assessment ratio study is also known as an assessment / sales ratio study.

    IDOR has documents describing the study and charts for each of the 102 counties in the state.

    Will find them and post the URL’s.

Leave a Reply

Your email address will not be published.