【正文】 or homes listed by brokers. Instead, it may be that this group of buyers would have paid a higher price regardless of whether or not they used an agent. This is a classic example of selfselection: If those who use an agent are predisposed to pay a higher price, the ordinary least squares estimate of the coefficient on the agent variable will overestimate the effect of the agent on the price paid by the buyer. The presence of selection bias could imply a number of possibilities. One is that the bias exists, and that after taking into account the selfselection by these buyers, housing prices are still higher for the group using a broker (as indicated by the coefficient of the broker variable). This result would imply that there are two separate residential real estate markets: one for brokerassisted properties, and the second for the forsalebyowner (FSBO) separation could permit the maintenance of housingprice differentials between the two groups of consumers. If selection bias is present, and housing prices are higher for brokerlisted properties, the price differences observed could be attributed to the predisposition of buyers using a broker to pay more for their homes. If no selection bias is present, the higher prices may represent the value added by the broker which the buyers are willing to pay for. Alternatively, no price differentials may exist due to the petitive impact of FSBOs on the prices paid for brokerassisted transactions. 6 The choice of whether to use a broker and its subsequent effect on price is modeled using a Heckman twostage model. In the first stage, a probit equation is employed to model the decision to choose a real estate broker. The decision is modeled as a function of eight variables. Buyer ine (AINC) is employed as a measure of the opportunity cost of search. It has already been shown that buyer search duration can be reduced by working with a real estate broker. If higher ine buyers have higher opportunity costs, we would expect that they would choose to buy through a real estate broker. Source: Leonard ,Harold ,Edward . Journal of Real Estate Finance and Economics. Academic Publishers,13:169181. 7 譯文： 通過(guò) 房地產(chǎn)經(jīng)紀(jì)人 決定 買房子 簡(jiǎn)介 房地產(chǎn)經(jīng)紀(jì)人究竟是做什么的，為什么購(gòu)房者會(huì)激 發(fā)他們的服務(wù)呢？在通常情況下，這是一個(gè)不容易回答的問(wèn)題。其實(shí)購(gòu)房者在通過(guò)經(jīng)紀(jì)人買房子的過(guò)程中買到的不僅僅是房子，而同時(shí)是在取得經(jīng)紀(jì)人的服務(wù)。 一些學(xué)者注意到買家對(duì)于那些業(yè)主代理上市的房子會(huì)支付更多。 第一步，買方?jīng)Q定是否聘請(qǐng) 一個(gè) 協(xié)助搜 尋房屋的服務(wù) 經(jīng)紀(jì)人 。 我們的研究結(jié)果表明，通過(guò)經(jīng)紀(jì)人買房的買家的確要比不通過(guò)經(jīng)紀(jì)人的買家支付更多的費(fèi)用。但很少有研究試圖衡量房地產(chǎn)經(jīng)紀(jì)人對(duì)住房市場(chǎng)的直接影 響，因?yàn)樵谌缃竦墓ぷ髦型鶗?huì)產(chǎn)生很多與研究文獻(xiàn)相矛盾的結(jié)果。根據(jù)來(lái)自北卡羅來(lái)納州的三個(gè)城市的住房交易數(shù)據(jù)，喬德認(rèn)為，經(jīng)紀(jì)人不影響他們出售的房屋的銷售價(jià)格，即使他們似乎影響了購(gòu)房者的住房消費(fèi)水平。 盡管 巴雷拉和祖帕諾 （ 1995）的首次研究沒(méi)有比較銷售價(jià)格，而是對(duì)經(jīng)紀(jì)人輔助交易和非經(jīng)紀(jì)人銷售交易進(jìn)行調(diào)查研究。 10 選擇性偏見(jiàn)的存在可能暗示了一些可能性。它已被證明，通過(guò)房地產(chǎn)經(jīng)紀(jì)人可以減少買方搜尋期間的工作。 出處： 倫納德 V這一結(jié)果將意味著有兩個(gè)不同的住宅房地產(chǎn)市場(chǎng)：經(jīng)紀(jì)輔助 為其中之一 ，如果選擇存在偏差，住房?jī)r(jià)格 的經(jīng)紀(jì) 屬性 就 越高，價(jià)格差異可能是由于購(gòu)房者傾 向使用經(jīng)紀(jì)人 來(lái)購(gòu)買他們的房子 。同時(shí)研究還表明，房地產(chǎn)經(jīng)紀(jì)人確實(shí)可以減少幾乎所有類型消費(fèi)者的買方過(guò)程的時(shí)間，無(wú)論是第一次買房的，買房經(jīng)驗(yàn)豐富的，本地的購(gòu)房者或者是外地的。同時(shí)也證實(shí)了 有經(jīng)紀(jì)人協(xié)助的買家市場(chǎng)要比非經(jīng)紀(jì)協(xié)助的買家市場(chǎng)大。 詹森和喬布森 (1980)在關(guān)于經(jīng)紀(jì)人選擇的第一個(gè)實(shí)證研究中發(fā)現(xiàn)，房地產(chǎn)經(jīng)紀(jì)人也會(huì)對(duì)價(jià)格產(chǎn)生影響。相反的，這類買家無(wú)論以何種方式購(gòu)買房子都會(huì)比其他買家花更多的錢。 8 第二步是購(gòu)買。較高的價(jià)格預(yù)示著買家相對(duì)于那些直接出售的房子更愿意較多支付那 些業(yè)主代理上市的房子。通過(guò)經(jīng)紀(jì)人購(gòu)買房子的一個(gè)好處就是可以減少搜尋房子的時(shí)間，從而降低了買房過(guò)程中的成本。 Jud and Frew, 1986) observe that buyers pay more for homes listed with brokers. It is not possible, however, to determine whether this applies to the entire residential housing market, since research to this point has relied on local data. Higher prices indicate a willingness