Minding Ag's Business

Appraisers vs. Robots

By Marcia Zarley Taylor , DTN Executive Editor
Some appraisers found big variations in property values with AcreValue, but their sample was small.

AcreValue, the new online farm real estate site I reviewed last week (see "Zillow for Ag: Farm Real Estate Dives into Digital Age" goo.gl/Hw0jU) is already packaging valuable information to renters and farm owners. Just remember the science is still a work in progress.

In 46 states, AcreValue offers field maps, cropping history and soil quality information on specific fields, derived from government databases. That's handy information that will help a prospective renter determine what to bid for cash rent. In four Corn Belt states (Iowa, Illinois, Indiana and Minnesota), AcreValue also offers robotic appraisals of individual farms down to their field boundaries. That's far more precision than you'd find with countywide land grant university or USDA land surveys.

Land quality is only one factor in AcreValue's formula; it also factors items such as location, interest rates, the county's tax environment and expected grain basis (see details at www.AcreValue.com ). Its statistical formulas can't yet account for irrigation or drainage improvements.

Sometimes certified appraisals on identical properties can vary radically from AcreValue's estimates, so the service is no substitute for hiring a pro when accuracy counts, several rural appraisers and realtors have told me.

For example, David Nebel, a certified appraiser and president of Hertz Appraisal Services, Nevada, Iowa, compared 35 properties to AcreValue's rough estimates. Nebel and his colleague Meghan Mills randomly sampled 35 cropland properties, not enough to be statistically valid but more of a anecdotal snapshot. They found AcreValue differed from Hertz's actual appraised values by 12% below appraised value to 56% above appraised value in some cases.

"AcreValue was closest on the average- and above-average quality farms, but exhibited significant error for the lower quality farms," Nebel says. "AcreValue can help start the conversation on values, but limits exist."

In response, AcreValue points out it only measures value on tillable acres. "We don't account for non-cropland or buildings that Hertz might be including in their appraised values. There could be other shortcomings to how they are comparing that doesn't make it true apples-to-apples," says Tamar Tashjian, general manager of AcreValue. There will always be outliers, but she stressed it would take a much larger sample to assess the service's accuracy, say 90% of the time.

I've also experienced wide variations when hiring professional appraisers to assess real estate values on the same property, so professional opinions can differ too. I asked Nebel if AcreValue differed much from the range you'd find if you hired two human appraisers.

"The general rule of thumb is that appraisers should be within 10% of each other," Nebel says."If the difference is more than 10% then something is 'wrong.' 'Wrong' as in miscalculations, errors in methodology or differences in assignment parameters.More times than not, if both appraisers are qualified agricultural appraisers and familiar with that market area, the difference is closer to 5%."

Bruce Sherrick, director of the TIAA-CREF Center for Farmland Research at the University of Illinois, points out that these online farm real estate valuation services are in their early stage. It's premature to test their accuracy, he argues; they will be refined over time.

"It's very healthy [for the ag real estate] industry, just like it was for the travel industry," he says. Only 15 or 20 years ago, you needed an agent to book your airfare, your hotel, your rental car. "Now I can plan a trip to Washington, D.C. in about one minute," he says, all online.

Give AcreValue a spin at www.AcreValue.com. I'd like to hear what you think of its accuracy and/or its usefulness.

Follow Marcia Taylor on Twitter@MarciaZTaylor

Comments

To comment, please Log In or Join our Community .

Pedro Sanchez 2/19/2016 | 10:24 AM CST
I have tried it. Looked at it. It is somewhat useful, but doesn't give value to anything that would be considered pasture or grazing/haying ground and building sites. I like the soil map part of it. It would be nice if we were able to build boundaries and get a value for an 80 out of a 160 acre piece.

The other thing I used it for was to give me a baseline for some auctions that were coming up. They valued most of the land around $6400, with some lower on these parcels. The sales ended up bring just under $4000 and just over $4000. Granted these were marginal pieces in my opinion. It also values my land around $8k/A. Land hasn't sold that high yet in my area, so I don't know how they come up with that figure.