February 2025 Blog: The State of Auction Estimates
Considering this is High Season in Florida, and the month of the “Extravaganza” Antiques Fair in West Palm Beach, it is surprising to see only one Highwaymen auction this month… at Manor Auctions.
This blog will cover 2 topics:
- My change in approach to the use of auction estimates
- The auction results ( data only) of the Manor Auction – primarily focusing on estimates.
In other words – this blog is a detailed assessment of the current state of auction estimates !
I. Auction Estimates:
When I started doing a statistical analysis of the Highwaymen auctions 2 years ago, I started with the target price being the midpoint between the Low estimate and the High estimate. This is the standard approach that all art businesses and art data websites use to evaluate the performance of sales at auctions.
For example, Mutual Art is the world’s largest art tracking service, keeping tabs on over 750,000 artists.
MutualArt tracks Highwaymen art, as well as the leading Highwaymen auction houses.
The methodology they use is to track sales based upon the midpoint between the low and high estimates. What they call the “mid-estimate”, rather than a “target price”. Thus, here is a screen shot from this service for two recent Harold Newton sales at Manor Auctions and Burchard Galleries, as recorded on the Mutual Art system.
In the Mutual Art system, performance is measured based upon the percentage above or below the Mid-Estimate.
When I started tracking Highwaymen results, this is what I did, as mentioned above. I received some criticism from the auction houses, but primarily from Manor, who stated that my approach was wrong for the following reasons:
- They said their target price for a painting at auction was the low estimate. They stated that the high estimate was not really an estimate – it was an aspirational price that did not reflect a “high” estimate.
- I was also told that the high estimate was an optimistic price that the consigner of the painting could hope to achieve. Essentially trying to convince consigner’s to consign their paintings because they are being told the value of their painting is within these estimates.
- ( Everyone should always remember that the true clients of any auction house are the consigners – not the buyers. The business of an auction house is driven by what they can get consigned. As long as they have consignments the items will sell 99% of the time. If they don’t have consignments, they don’t have a business. )
As a result of these discussions, I changed my approach to use the low estimate as the target price for Highwaymen auctions. During the last two years I have found the estimates have become more and more inaccurate, with a typical auction having 40% of the paintings not even reaching the low estimate. This is an incredible disservice to new collectors who do not have the experience to know painting values and rely on the auction houses to provide accurate estimates of the value.
Having done my analysis for the last two years, I have come to the conclusion that auction estimates have become unrealistic and inaccurate measures of what a Highwaymen painting might sell for. I also know that my blogs have put pressure on the auction houses to bring their estimates down to reality, but it still has not been enough.
As such, I have decided to go back to the art industry standard in the more widely based art market, and will use “mid-estimate” as the basis for how I assess the results of an auction beginning this month, and going forward.
II. The February Manor Auction
First, let me say that beginning with this market review, I do not plan to discuss the Manor auctions, other than summarizing the statistical performance. The data is the data, it is all publicly available, and Manor even publishes it themselves in their own website and now in their own auction blogs. However , following more than a year of legal threats from Manor about the content of my blogs (which have generally been positive about Manor), I have decided to not discuss Manor’s activities going forward. Personally, I feel if I discuss and promote the other auction houses, and not Manor, it is their loss – not mine. I will limit my comments to the statistics.
Basic Statistics:
The February Manor auction had a strong crowd I was told by several people attending, but had only 75 viewers online on LiveAuctioneers, and an average of 95 on the Manor+Invaluable website. This is a relatively small on-line viewer base, as auction houses like Vero Beach Auctions typically average about 450 online bidders. The auction was not streamed live with video this time, as it was in January. Apparently, there was no mention of this in the auction room.
The auction contained 70 Highwaymen paintings and 3 Backus paintings.
Mid-estimate values for the Highwaymen paintings were $285,450, and hammer prices totaled $213,450. Realized hammer prices were 25% below mid-estimates. At the same time, the number of paintings not reaching the mid-estimate was 57, or 81% of the lots offered. This suggests that low estimates were largely reached but with 81% of all lots not reaching the midpoint, it clearly leads to the conclusion that high estimates were far higher than they should be.
Of the 3 Backus paintings, none reached the mid-estimate. Mid-estimate totals were $85,000 with total hammer prices at $64,000 – a serious shortfall from the estimates.
This is actually consistent with the comments I made earlier in this blog that high estimates are knowingly set well above where they should be.
Manor Estimates:
As all Highwaymen auction-goers know, the typical estimate at Manor looks something like $2,000-$4,000, $3,000-$6,000, or $5,000 – $10,000. Almost always the estimate has a spread of 100%. For an auction house that should know the market well, this is an extremely wide spread, and is out of synch with the standards in the art industry. Art auction estimates are typically much more narrow in their spread between low and high estimate, as they should be. The underlying message in an estimate of $5,000-$10,000 is that the auction house doesn’t really know what the painting should sell for.
Recently, it seems that Manor has been lowering their overall estimates following the 4th quarter of 2024 where frequently a majority of paintings did not reach the low estimate. The auction this month had some strange estimates, indicating that the auctions estimates are also not being done in a consistent manner.
For example, take the estimates on the Harold Newtons, perhaps the most well known Highwaymen artist, and one with always a large number paintings being offered. This artist should be the most easy to develop an accurate estimate for the value of a painting, when the auction house sells at least 6-8 paintings by Harold every month.
In February the Harold Newton offerings were as follows:
| Lot No. | Low Est. | High Est. | Hammer Price |
|---|---|---|---|
| 16 | 8,000 | 16,000 | $6,600 |
| 17 | 7,000 | 14,000 | $5,800 |
| 18 | 7,000 | 14,000 | PASSED |
| 19 | 6,500 | 8,500 | $5,400 |
| 20 | 6,000 | 8,000 | $5,200 |
| 21 | 5,000 | 7,000 | $4,200 |
| 22 | 4,000 | 8,000 | $6,200 |
| 23 | 3,000 | 6,000 | $3,200 |
For the artist who is probably the most well understood in terms of price history and pricing, this set of 8 paintings tells a story of either lack of knowledge, or incredible bad luck.
- 5 of the 8 paintings have a 100% spread from the low to the high estimate. This immediately should tell a collector that the estimates are not accurate because there is not a clear understanding of what the painting is worth.
- Despite the wide 100% spread in estimates, 7 of the 8 paintings did not even reach the low estimate, clearly indicating the estimates were artificially high across the board. How could any auction house get 7 out of 8 wrong – unless they are trying to convince bidders the values are higher than the painting is really worth.
- Lots 19,20,21 had a much smaller spread in the estimate range – which indicated a better understanding or view of where the painting should sell. Nonetheless, those estimates were all still too high as well.
Another example were 2 Sam Newton paintings – lot 63 and lot 64, with estimates of $3,500-$7,000 and $3,500-$5,500 respectively. How could one painting have the apparent Manor standard of a 100% estimate spread, and the next painting have the same low estimate with a much lower high estimate. In the end, the estimates didn’t matter here either because both paintings failed to reach their low estimates.
Conclusion: Bidders should place no value in the estimates the auction house is ascribing to the paintings being auctioned. Or – bidders should always assume the true value is about 15-20% below the auction estimates.
Next Month:
Next month will be the last big full month of auctions for the winter season.
There will be 4 major auctions focused on Highwaymen paintings, starting with Florida Highwaymen Auctions on March 8. There will also be a Vero Beach Auction – date not yet determined. And the month will conclude with what is always the blockbuster sale of the major auction months – Rennick Auctions on Sunday, March 23.
Wishing everyone success in their Highwaymen Hunting !
Mark Lerner