More Data on Real Estate Agents

Redfin is a discount real estate brokerage in Seattle that has put together some interesting data analysis from their first year in business.

Their numbers suggest that clients who use their discount brokerage firm pay a lower percent of the list price than the typical home buyer in Seattle (99.329% of the listing price with Redfin vs. 100.233% with other agents). That difference in price is worth more than $4,000 per purchase, on top of the already discounted fees/commission sharing.

Of course, it is hard to know exactly what to make of these results. Redfin would like you to believe it is that they have crack negotiators doing the bargaining. Another possibility is that the home buyer that seeks out a discount broker is cheaper and more savvy than the typical home buyer to begin with. It is possible (although there is no direct evidence to support such a claim) that these homebuyers could have made even better deals if they had worked with a full-service agent.

It is not easy doing data-driven research that is completely convincing. This research is not completely convincing. I do think it is sensible and interesting, at least, so hats off to Redfin.

Although the data here do not show it, one of the legitimate fears a client should have when using a discount broker is that lower fees translate into even weaker incentives to do well for the client. This is why I think that the real-estate market is more likely to move towards a flat-fee system (where a fee is paid to list houses and then agent compensation is based on an hourly wage as opposed to a share of the sale price), rather than to a discount system which is still tied to sales price.


"compensation is based on an hourly wage" - beware the law of unintended consequences! I can just see the levels of busy-work that would erupt. Think about lawyers. Every time your real estate agent called you, another (rounded up) 30 minutes of billable time... For buyers, they would have an incentive to encourage you to keep looking. For sellers, they would price high and so it would take longer to sell.
You can argue that the market would settle it out, but has not really happened with lawyers and comparative analysis is hard with specialist services such as home buying/selling (every case is different).
Flat fees make more technical sense, but then you run the risk of an agent that wants to push the sale even faster (they get paid the same regardless of whether you were screwed or not).
The only thing I can see that would work is bonuses based on exceeding performance criteria set by the customer. You still have the agent making the "is it worth X extra dollars" calculation, but you can choose the terms.



Since regular commissions are stated as a percentage of selling price, it's possible that the after-sales-cost price of a home received by a seller is the same whether using a regular agent or a discount agent (indeed, the theory of competitive markets suggests this would be true.

Second, the difference does not take account of the differential effort made by sellers (not their agents, the sellers themselves) when using alternative agents. If sellers using dicsount brokers have to do more work, then we'd expect their after-sales net to be greater.

Third, there's time-to-sale--under which representation arrangement do homes sell faster? A quicker sale at a lower price might be a larger discounted present value.


Redfin states that their agents are paid a salary. So although the business collects a discounted percentage fee, the agents are paid a fixed salary plus a bonus determined by customer satisfaction.

"This is why we decided to pay Redfin agents a salary with a customer satisfaction bonus, not a commission. Agents do what you pay them to do, we reasoned, and we believed our agents would be more likely to get the price our customers wanted."


What drives the far higher commissions that US home sale transactions attract compared to the UK? Here in the UK, the seller pays between 1% and 3% to their agent for finding a buyer. The buyer pays nothing (except in rare cases where the buyer also retains an agent, this only happens at the very high end of the property market, presumably because people spending many millions on a home would rather pay someone to find it for them).

In the US it appears the seller pays 3% to their agent and another 3% to the buyers agent, is that right?? A total of 6%? Why is it so much more expensive to sell a home in the USA? Are agents earning twice as much, presumably not. Maybe there are just too many agents, so they each close fewer deals and so need to charge more per deal? Anyone know?


First of all, we are deeply honored to be on the Freakonomics blog. We agree with Steven that one factor working in Redfin's favor is that our customers tend to be already price-sensitive, which we acknowledged in the original analysis.

But our main rationale, which we believe makes the data convincing rather than simply interesting, is that we pay our agents a salary and a significant customer satisfaction bonus, not a commission. The bonus is paid based on customer responses to an online survey, distributed to everyone who tries to do business with us.

How you compensate agents directly affects what they're motivated to do. A commission, whatever its size, motivates the buyer's agent to close at any price; the agent earns more when the buyer pays more. A customer satisfaction bonus motivates agents to get the price customers want.

Whether commissions motivate the agent to get a higher price or a lower price, we agree with Steven that a discounted commission naturally reduces the capacity to provide good service, but with one important caveat: that technology can yield efficiencies allowing for better service at a lower price.

This caveat is important. Steven's analysis of Redfin is based on the assumption we are a discount broker, which attempts to provide the same service as a standard broker but at a reduced price. We are no more a discount broker of real estate services than, Expedia and E-Trade are discount brokers. We are an online broker; we provide a different service at a reduced price.

Our cost structure and our business model are radically different. Redfin has virtually no customer acquisition costs except from development of our website; traditional agents spend 67% to 90% of their time looking for new clients.

We can afford to pay our agents double or triple the industry average, so we get the cream of the crop, and those agents can focus exclusively on closing deals; customer find the home they want to buy on their own.

We directly employ our agents, allowing us to manage quality. We can offer a 100% customer satisfaction guarantee, which is impossible in a traditional brokerage with franchises and independent contractors. Our agents own equity in the company, so they tend to think long-term, and they tend to work together.

This rationale may or may not be convincing. Rationale aside, we would ask: what data would be convincing? What data supports the conventional wisdom, that traditional real estate agents' high commissions are offset by negotiating savvy?



Redfin issues a press release with unprovable figures that they claim they cannot release due to MLS rules.

A quick call to the local MLS confirms that use of these statistics ARE allowed.

End-of-year stats from the MLS:

Scroll to Page 27.

View "King County"

Median List Price: $520,797
Median Sold Price: $425,000

Buyers paid an average of 81.61% for all single family homes in King County for the year 2006.

If Redfin customers are paying 99.329% of the list price for a home, in roughly the same 12-month period (Jan-Dec v.s. Feb to Feb), then they may be paying a great deal more than they need to for a house.


We were getting outrageous buys for our elated San Diego purchasers in 2006. Much of this success came from excellent agent relations--and information that was not listed on the MLS.

Certain information can never be publicly posted (at seller's request); rather, it is shared discretely between agents (with seller's permission). Said information could relate to reasons for a seller's motivation in selling (divorce, financial difficulties, transfer, etc).

This is an area where technological efficiencies fail.


Isn't there a commission structure that would make more sense for sellers and their agents? Instead of (for example) 6% of full selling price, make it 0% for the first 90% of the asking price, then 60% of every dollar above that 90% level. Thus if the sale is made at full asking price, the agent is making the same as they were before, but now the incentive to fight for the extra $1000 is 10 times greater than before.

The one caveat is that the seller must make sure the asking price (and the 90% starting point) is set at a place they are comfortable with.

As for buyer's agents... well having one of those never has made sense to me.


Completely agree with the author. In fact I quote your book in my book Commissions at Risk. In it, I argue that flat fee and low cost brokers will be THE way transactions are handled in the future. I am a technologist; our goal in technology is to create efficiency and market transparency is required to get there. Every other industry has gone this way, with the exception of the "high end" in each business. How many people still call a travel agent? A real estate agent? Many of the travel agents were given jobs by online companies and paid salaries to do the same work. Real estate is going to take longer thanks to the ten-decade old MLS and the inherent protectionist system that is set up, but it's only a matter of time before this industry becomes competitive thanks to technology.


The 81% you quote seems wrong to me. As a resident of King County, WA, I've never heard of a house in this area selling for 20% below asking price unless something catastrophic had happened to it (i.e., it burned to the ground after it was listed on the MLS).

Maybe we could roughly validate those numbers by extrapolating from your performance over the last year.

On average, did all of the sellers you represented in 2006 only receive 81% of the original list price?

When you were a buyer's rep, was your average client successful in purchasing the house for 81% of the original list price?

I'm not expecting an answer. We both know that 81% number is baloney.

By the way, real estate agents like to argue that they deserve a 6% commission because their knowledge of the market enables them to extract more value for the seller than a discount broker. What does it say about their argument if Marlow's number is true and every seller's agent in Seattle is over-pricing by an average of 25%?




An Agent working for an hourly wage? Not gonna happen.

The fact is that this system is screwed up but rather than trhow the baby out with the bath water, the industry could do its own policing by making it harder to become an agent.

I recently interviewed one of the first flat fee and variable fee agents in the Country.
Real Estate Experts


Steven Levitt,
"Redfin would like you to believe it is that they have crack negotiators doing the bargaining. Another possibility is that the home buyer that seeks out a discount broker is cheaper and more savvy than the typical home buyer to begin with. It is possible (although there is no direct evidence to support such a claim) that these homebuyers could have made even better deals if they had worked with a full-service agent."

It's also possible that the moon is made of blue cheese (although there is no direct evidence to support such a claim).

If I understand your thought fragment correctly, you're implying that Redfin's buyers are more savvy and get better results than the buyers that go to traditional brokers. In essence, that Redfin is getting the best deal-makers as clients and therefore comparing the Redfin clients to those of traditional agents isn't an apples to apples comparison so Redfin's analysis isn't accurate.

Think about the (lack of) logic behind what you're implying. On the one hand, you're crediting Redfin's clients with being so good at deal-making that it's not fair to compare them to those of a full-price broker. On the other hand, you're saying that Redfin's clients might be able to negotiate a better result by using a full-service broker and by implication aren't savvy enough deal-makers to choose the broker that will result in the best deal. Which are they? Super-savvy consumers or dopes who could have gotten a better deal with a traditional broker? If the latter, what does it say about Redfin that their "dopey" clients got a better result than the "smart" clients that chose a traditional broker? If the former, what does it say about Redfin that the super-savvy deal-makers choose Redfin?

Your comment seems to me like a gratuitous slam on Redfin, without any real thought behind it. Given your reputation, I expected better from you.



I would say from experience very little is really known about the inner workings of the typial real estate brokerage. Money is the number one ranking system and trumps "cleint care" or professionalism in almost every case. Why not consider the idea of a satisfaction guarantee, hold agents to a higher standard and fill a niche as a "discount" broker. How this is considered a discount when we live in a market oriented society is beyond me. It is a sham the government considers agents not to be employee's with all the control a broker exerts. Is it not true you have to use broker owned services or see a reduction in income, advertising on "your" listing?
The broker owns all listings yet will penalize an agent who fails to toe the line. IRS is not blind, yet it is due to powerful PAC's this is not addressed.
Follow the money is all one needs to say. If the agent was a true independent contractor and followed all the laws, it would be a different story. Employee status would ensure broker liability, SDI taxes would be fairly apportioned and fewer agents would be needed. I think we all know this is true but will anyone listen. Not if money is lost. Fewer agents also means less income to the national and state boards, loss of perks to the diva's who run the MLS and boards. Oh I should write a book.. and I haven't got to the lenders yet. Go Redfin. I already tell many people about the site and welcome it to the market place. I just hope they know how to keep the good agents when the market gets better.



I forget the negotiations part of the above argument. How many offers has a typical buyers agent been even allowed topresent. In the 1970's we were able to present in person, with very few exceptions.
Today I never have the "right" as every seller seems to not want that or the agent tells them it is not wise. I have to fax or email an offer and unless I have a personal relationship with the agent, never see them and have to deal with a transaction coordinater during escrow. Many buyers agents also refuse to present offers and some of that is why go there and be asked to step outside for an hour or more while a coounter is being drawn up. So the negotiations, in my area of CA, Orange county.... is minimal at best and one sided. Unless you work in the same office and present no challenges to the brokerages "services" ( escrow, title and loans) you will be shut out. I also see RESPA violations in the counter
offers regarding buyers title 100% of the time. Yet no one will listen to a complaint and it is just understood that the money is the issue, not the client. I challenge anyone to make a case this is not true.



tommac: We negotiate every step of the way on behalf of our clients--even down to escrow fees. For example, the other agent may choose services for their listing, but we stipulate in the offer that escrow fees may not exceed $150 plus $1.50 per thousand up to a million--or $100 plus 1.00 per thousand over a million (and negotiated perhaps even lower if much above a million). We also look over loan terms and watch out for superfluous fees.

An important asset in being a crack negotiator is having strong relationships in the business along with the ability to sometimes work out terms for an offer before it is ever written.

We are working with a client now who is purchasing a home under short sale conditions. It has taken time, patience and lots of hard work--but it appears she willl be able to purchase the $699,000-appraised home in Carlsbad, CA for $567,000. In fact, a model match across the street was just listed at $699,000.

These terms were never in MLS and the offer would never have worked if the listing agent didn't know and trust us--and if we hadn't volunteered to assist with lender negotiations. Meanwhile, the listing agent is working on behalf of her seller to lessen his financial liability.

These are just some of the things crack negotiators do in a buyer's market. And when the cycle turns back to one that favors the seller, it will once again pay to have trusted friends in the business.

Why? Because our buyers love to hear whispers of listings that have not yet hit the MLS and where no sign has been posted. And because we work well with other agents, we can sometimes get our clients in to see these hot properties before or during the staging process.

Do we earn our money? Absolutely.

Is is a prime motivator? What a bore.



Buyers Agents should charge similar to the attorney or accountant hourly model:
Interview and asses the client and their needs.
Determine the amount of hours it should take to handle their situation.
Establish a retainer contract that agrees to pay the BA this cost POC.
If the scope of work obviously exceeds the retainer $$, start adding up the hours and charge appropriately.

Buyers Agents Beware


Some legal clients cannot afford/do not want the retainer model of business and prefer one based on success contingency.

Ditto for real estate.

And come to think of it, I wouldn't mind my accountant working on a contingency basis, either. Somehow, I think he might become a more aggressive tax planner.

Finally, what happens if an agent agrees to work on a retainer basis with fees paid at close of escrow--and the transaction blows out of the water? Most buyers don't want to pay agents for failed transactions.


We are a flat fee brokerage on both sides of the transaction. No dual agency, no referral fees, every property equal in value to us, every negotiation made based on how much the client likes the property and is willing to pay. They aren't talked into paying more because somebody next door did. Every client has loved the service and felt they got more than their money's worth. Have had client's tell me they feel bad and want to pay more on bigger commissions. Can't do that, it's liberal. Charging someone more because they have more money when the amount of work is the same. That is like taxing the rich because they have higher incomes. I would only charge more if I had to do more work. I started this business model with no venture capital investors telling me I have to pimp out possible way to take money from clients. We do our job to make our client's happy and in turn they will refer our services to others. We do not lose money on listings ($2000 Refin) so we can use those listings to capture prospects for the selling office commission. I'm proud of this business model. Had a realtor tell me that I wasn't trying to make any friends in the industry and I replied, "Sure I am. I'm trying to make friends with the buyers and sellers".



I know the perception is that RE agents go for the highest price because they will earn more on commission. That is not my experience. I have recently on a couple of transactions based my commission on a different/lower price.

The circumstances were that the buyer's were offering higher prices than either the list price or a mutually agreed upon price (in one case I represented the buyer and in another the seller) and I (with the other agent's approval)changed the contract to reflect that the commission for both the seller's agent, and the buyer's agent would be based on the lower price, and not the actual selling price.

It was not in our client's best interest to pay for the higher actual selling price, when the higher price was due to negotiated terms that benefited both clients.

This would also be an example of why the statistics on these particular transactions would not reflect the true value the client's recieved because of negotiated terms that are not reflected in the NWMLS data. It is the uniqueness of every transaction that makes the data generated to the NWMLS so difficult to measure in the terms Redfin has published.



Real estate is not an efficient market!

As a very happy Redfin customer I am grateful for what they are doing for consumers but I contend there is a much bigger problem no one seems to be talking about. Unlike the stock market where you can assume all information is priced in to a security, housing information is not freely available and is horded by agents and the MLS. The reason is obviously to protect their precious 6% commission but what will happen to housing prices when Redfin is successful in changing the real estate market in it's entirety?

Selfishly I feel like I got a GREAT deal on my house because I was able to take advantage of technology, information, and marketing timing, to buy at a much lower price than if I used a traditional agent. However, I am worried that when I sell my house some time in the future all the agents propping up the market will be gone and we won't be able to sustain the kind of gains we have enjoyed in the inefficient market.

Steven, are agents actually good for the economy because they create false value in housing prices?

Matt B.