Welcome to City-Data.com Forum!
U.S. CitiesCity-Data Forum Index
Go Back   City-Data Forum > General Forums > Real Estate
 [Register]
Please register to participate in our discussions with 2 million other members - it's free and quick! Some forums can only be seen by registered members. After you create your account, you'll be able to customize options and access all our 15,000 new posts/day with fewer ads.
View detailed profile (Advanced) or search
site with Google Custom Search

Search Forums  (Advanced)
Reply Start New Thread
 
Old 09-24-2012, 04:15 PM
 
Location: Orange County, CA
204 posts, read 338,311 times
Reputation: 95

Advertisements

Quote:
Originally Posted by lvoc View Post
You are suggesting replacing one system that does not reward good performance with another somewhaat more complex one that does not reward good performance.
Is x% * sales price a good compensation scheme for a seller's agent? Agree or disagree, and explain why.

Great quote, by the way.
Reply With Quote Quick reply to this message

 
Old 09-25-2012, 09:58 AM
 
Location: Columbia, SC
10,965 posts, read 21,991,425 times
Reputation: 10685
Price fixing, really? Most every place I go I can buy a coke for a dollar. Pepsi also cost about a dollar. Are they price fixing also?

Also, in regards to your question: You want to know why a 5% of 100,000 is better for a consumer is better than a $7500 flat rate? Because in the 5% model the consumer pays $5000 and in the flat rate the consumer pays $7500. I'm assuming the same service is rendered is equal so the % is better because it would have saved the consumer $2500. Does that explanation work?
Reply With Quote Quick reply to this message
 
Old 09-25-2012, 02:46 PM
 
Location: Orange County, CA
204 posts, read 338,311 times
Reputation: 95
Quote:
Originally Posted by Brandon Hoffman View Post
Price fixing, really? Most every place I go I can buy a coke for a dollar. Pepsi also cost about a dollar. Are they price fixing also?
Different outlets charging the same price for the same product manufactured by the same company isn't quite the same. If 11 different sodas charged exactly the same price, that'd be a better comparison. It probably happens more often in retail than you might expect (there was a case involving CDs and Wal-Mart a few years back), but for such low cost items, there isn't much political pressure to investigate.

In general, there is a well-known "concentrated benefits, distributed costs" problem. Consumers are less organized and concentrated than producers, which is why tariffs are common even though the overall effect is to harm the economy. Anti-trust isn't investigated nearly as often as it ought to be (and I think the Microsoft case was wrongly decided), so whether or not the government does something isn't much of an indicator.

Still, my point was never on illegality, but identifying a practice that could be improved upon. As I see it, pricing standards impede price competition. I'm still waiting to hear what purpose such a pricing standard serves for the buyer.

Quote:
Originally Posted by Brandon Hoffman View Post
You want to know why a 5% of 100,000 is better for a consumer is better than a $7500 flat rate? Because in the 5% model the consumer pays $5000 and in the flat rate the consumer pays $7500. I'm assuming the same service is rendered is equal so the % is better because it would have saved the consumer $2500. Does that explanation work?
No, because "flat rate" does not necessarily mean a higher rate. It refers to the compensation plan where the rate is agreed upon between the buyer and the agent and is set to a fixed number before the house search and house price negotiations. That number could very well be $4,000 or $3,500 or $8,000. Whatever number it is, it is written into the buyer rep agreement and does not change regardless of the price of the house. It is payable upon close.

It avoids the conflict of interest with x% * sales price for buyer's agents and is also simpler (buyer's rep just lists a number -- no formula) and many agents have already been using it for a while (perhaps not exactly as I outlined above, I'm not sure).

It fails to align incentives like x% * sales price accomplishes for listing agents (thus my unanswered question in the previous post).

Last edited by perfectlyGoodInk; 09-25-2012 at 04:14 PM..
Reply With Quote Quick reply to this message
 
Old 09-26-2012, 10:51 AM
 
Location: Columbia, SC
10,965 posts, read 21,991,425 times
Reputation: 10685
So you will not admit that a consumer paying $5000 for the same service comes out ahead compared to $7500? Then this conversation cannot go further as you will consider things with an open mind if you cannot even admit that simple truth.
Reply With Quote Quick reply to this message
 
Old 09-26-2012, 11:27 AM
 
Location: Orange County, CA
204 posts, read 338,311 times
Reputation: 95
Quote:
Originally Posted by Brandon Hoffman View Post
So you will not admit that a consumer paying $5000 for the same service comes out ahead compared to $7500?
Flat rate is not defined as "$7,500." It's when the commission is set to a fixed number, agreed upon before the home search and negotiations. That number could be $5,000. According to your argument, if the flat rate is $7,500 and the home price is $300,000, then flat rate is better because $7,500 < $9,000. What you have done is the rhetorical equivalent of saying, "a blue car is a faster car than a black car because when a blue car is going 80 MPH and a black car is going 50 MPH, then 80 > 50" (ignoring that the black car can also go 80 MPH and a blue car 50 MPH).

This also does not take into account that the service of the buyer's agent is finding and closing a house that is a good fit for the buyer -- for as low a price as possible. The house price is not known at the beginning, so an agent costing $7,500 could actually very well be worth that extra $2,500 if they found and closed an equally valued home for $25,000 less than the $5,000 agent would have (remember not to confuse price with value -- for houses or agents).

A compensation plan should not be judged solely on cost to the buyer, or else a flat rate of $0 would obviously be the best. Instead, it should be judged on whether it rewards or punishes agents for performing well. The standard of x% * sales price punishes agents for the "as low a price as possible" part of good performance. Flat rate neither rewards nor punishes it (but it is the simplest). An inverse formula of x% * ($YYY,YYY - sales price) rewards good performance. None of this is surprising when you acknowledge the simple truth that x% * sales price was designed for listing agents, not buyer agents.

Last edited by perfectlyGoodInk; 09-26-2012 at 12:41 PM..
Reply With Quote Quick reply to this message
 
Old 09-26-2012, 12:49 PM
 
12,973 posts, read 15,807,980 times
Reputation: 5478
Again though you are pushing a theoretical argument against a practical reality.

Given the result is revenue neutral their is no particular gain for the buyer. The result needs to be revenue neutral to maintain the existence of the buyers agent.

Neither actually rewards good performance. They simple utilize different mechanisms to get to the same end

The problem that you are working and have not solved is that the value is, for practical purposes, unknowable with enough precision to be useful. Thus an actual measure of performance is not very practical. And the system you suggest would, if implemented, simply be gamed in a different way.

As Machiavelli pointed out changing systems is difficult and dangerous. You need something with big rewards. This is not it.
Reply With Quote Quick reply to this message
 
Old 09-26-2012, 02:32 PM
 
Location: Orange County, CA
204 posts, read 338,311 times
Reputation: 95
Quote:
Originally Posted by lvoc View Post
Given the result is revenue neutral their is no particular gain for the buyer. The result needs to be revenue neutral to maintain the existence of the buyers agent.
Yes. All three compensation plans have variables that can be easily modified to remain revenue neutral. Since the agent is the one to prepare the paperwork, they have the most leeway in deciding what compensation options appear on the buyer's rep and the actual numbers used. The agent can even present the terms as non-negotiable, and this is fine in regards to market competition and efficiency because the buyer is also free to walk away and find somebody else if they don't like the terms.

Quote:
Originally Posted by lvoc View Post
Neither actually rewards good performance. They simple utilize different mechanisms to get to the same end
Here's how you defined performance earlier:

Quote:
Originally Posted by lvoc View Post
Performance is finding the correct house for the client and then buying it at as low a price as possible.
Note that there are two key parts of this: 1) "correct house" and 2) "as low a price as possible. By "correct," I presume you mean a home that is as highly valued by the client as possible (there isn't just one single correct house for a buyer, or else the job would be nigh impossible).

Note, economics defines consumer surplus (the gain to the consumer from an exchange) as the difference between the value of the good/service purchased and the price paid by the consumer for it (value - price). Value and price are actually never equal. The value must be higher than the price for the buyer to be willing to pay the transaction costs to obtain the good/service. Now, there are two ways to maximize this gain: either increase the value (the first part of agent performance), or decrease the price (the second part of agent performance).

None of the three compensation plans uses the commission amount to reward the value side of agent performance. As you say, this is unknowable and not easily inserted into a formula. Instead, all three compensation plans -- including the standard one -- rely upon the contingency of the payment as well as referrals to provide incentives for the agent to find a house the buyer likes and is happy with. They use the same mechanisms, not different ones, and I don't think this part can be improved upon.

However, there is still the "as low a price as possible" part of agent performance. The standard x% * sales price penalizes this performance. Flat rate is neutral to it. x% * ($YYY,YYY - sales price) rewards it. More importantly, reducing the price increases the gain to the consumer just as much as increasing the value (remind them of what else they can buy with that money -- granite countertops, a car, college savings, etc.). Given how expensive houses are, an agent saving a buyer tens of thousands of dollars dwarfs the amount of the commission itself (i.e. 3% is a small percentage of the savings). That is the gain to buyers, not lower commission. If their goal is to minimize commissions, there's RedFin or buying unrepresented.

The reward to agents is market share. In the Information Age, it is pretty much inevitable that consumers will eventually figure out the above (many have already). If you are ahead of the curve, you can win them from other agents and from the Internet. Successfully saving your clients lots of money should help your referrals -- without the commission penalty you would bear under the standard commission plan.

Last edited by perfectlyGoodInk; 09-26-2012 at 03:19 PM..
Reply With Quote Quick reply to this message
 
Old 09-26-2012, 03:35 PM
 
12,973 posts, read 15,807,980 times
Reputation: 5478
Quote:
Originally Posted by perfectlyGoodInk View Post
Yes. All three compensation plans have variables that can be easily modified to remain revenue neutral. Since the agent is the one to prepare the paperwork, they have the most leeway in deciding what compensation options appear on the buyer's rep and the actual numbers used. The agent can even present the terms as non-negotiable, and this is fine in regards to market competition and efficiency because the buyer is also free to walk away and find somebody else if they don't like the terms.



Here's how you defined performance earlier:



Note that there are two key parts of this: 1) "correct house" and 2) "as low a price as possible. By "correct," I presume you mean a home that is as highly valued by the client as possible (there isn't just one single correct house for a buyer, or else the job would be nigh impossible).

Note, economics defines consumer surplus (the gain to the consumer from an exchange) as the difference between the value of the good/service purchased and the price paid by the consumer for it (value - price). Value and price are actually never equal. The value must be higher than the price for the buyer to be willing to pay the transaction costs to obtain the good/service. Now, there are two ways to maximize this gain: either increase the value (the first part of agent performance), or decrease the price (the second part of agent performance).

None of the three compensation plans uses the commission amount to reward the value side of agent performance. As you say, this is unknowable and not easily inserted into a formula. Instead, all three compensation plans -- including the standard one -- rely upon the contingency of the payment as well as referrals to provide incentives for the agent to find a house the buyer likes and is happy with. They use the same mechanisms, not different ones, and I don't think this part can be improved upon.
So we agree here. Actually I often don't know what the actual commission is. My partner does but she would only tell me if it was something unusual - under 2% or 0% which sometimes does happen.

Quote:
However, there is still the "as low a price as possible" part of agent performance. The standard x% * sales price penalizes this performance. Flat rate is neutral to it. x% * ($YYY,YYY - sales price) rewards it. More importantly, reducing the price increases the gain to the consumer just as much as increasing the value (remind them of what else they can buy with that money -- granite countertops, a car, college savings, etc.). Given how expensive houses are, an agent saving a buyer tens of thousands of dollars dwarfs the amount of the commission itself (i.e. 3% is a small percentage of the savings). That is the gain to buyers, not lower commission (if they want to minimize commission, there's RedFin or buying unrepresented).

The reward to agents is market share. In the Information Age, it is pretty much inevitable that consumers will eventually figure out the above (many have already). If you are ahead of the curve, you can win them from other agents and from the Internet. Successfully saving your clients lots of money should help your referrals -- without the commission penalty you would bear in the standard commission plan.
Ahh but back to number one. Any such formula is subject to gaming. The Agent is at least tempted to manipulate the value to minimize the cost. You give him money to sub-optimize in a specific direction. And some will.

I would also point out that the consumer is not really in much of a position to negotiate against an experienced agent. An experienced agents are exactly those who take advantage of the consumer...and most interesting the consumer often loves it. I don't know how many times I have seen a seller who was clearly taken advantage of claim what a great agent she/he had on the other end.

Your belief in the intelligent buyer with copious information is also misplaced. There are some and dealing with them is also a pleasure even though there may be some sharp discussion about commissions. On the other hand even the very intelligent are often not very good and have to be led by the hand.

Changing the system is tough. Requires clear and obvious advantages. You have not yet gotten to a clear break even...anymore an advantage.
Reply With Quote Quick reply to this message
 
Old 09-26-2012, 03:39 PM
 
Location: Orange County, CA
204 posts, read 338,311 times
Reputation: 95
Quote:
Originally Posted by lvoc View Post
Ahh but back to number one. Any such formula is subject to gaming. The Agent is at least tempted to manipulate the value to minimize the cost. You give him money to sub-optimize in a specific direction. And some will.
Yes, all formulas are subject to gaming. For example, how would one game x% * sales price? An agent tempted to maximize their commission would do what? What can an experienced agent under this commission plan do to take advantage of a client who is not-so-intelligent?

Last edited by perfectlyGoodInk; 09-26-2012 at 04:47 PM..
Reply With Quote Quick reply to this message
 
Old 09-26-2012, 05:01 PM
 
12,973 posts, read 15,807,980 times
Reputation: 5478
Quote:
Originally Posted by perfectlyGoodInk View Post
Yes, all formulas are subject to gaming. For example, how would one game x% * sales price? An agent tempted to maximize their commission would do what? What can an experienced agent under this commission plan do to take advantage of a client who is not-so-intelligent?
Agents upsell all the time. It is a problem. But at least it runs into the normal response of affordability. Down sell might be an even bigger problem as the natural response is to favor spending less money.

Under any likely outcome it does nothing to optimize the situation.

To sell what you are trying to sell you need big advantages. You don't have any.
Reply With Quote Quick reply to this message
Please register to post and access all features of our very popular forum. It is free and quick. Over $68,000 in prizes has already been given out to active posters on our forum. Additional giveaways are planned.

Detailed information about all U.S. cities, counties, and zip codes on our site: City-data.com.


Reply
Please update this thread with any new information or opinions. This open thread is still read by thousands of people, so we encourage all additional points of view.

Quick Reply
Message:


Over $104,000 in prizes was already given out to active posters on our forum and additional giveaways are planned!

Go Back   City-Data Forum > General Forums > Real Estate
Similar Threads

All times are GMT -6. The time now is 09:44 PM.

© 2005-2024, Advameg, Inc. · Please obey Forum Rules · Terms of Use and Privacy Policy · Bug Bounty

City-Data.com - Contact Us - Archive 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37 - Top