The Impact of Mergers on the Value of Merging Firms’ Suppliers and Customers

1. Introduction

Our research intends to examine the impact of mergers, especially the horizontal mergers on the value of merging firms’ suppliers and customers. We make use of the event-study method to do the research. In this report, we will show the results and give our explanation of the results.

2. Results

The results of our research are exhibited at Table1 and Figure1.

As to the bidder’s customers and suppliers within the full sample, the estimated means of the cumulative abnormal return on intervals of [-1, 1], [-2, 2] and [-5, 5] are all negative, which is consistent with our assumption that the return of the companies which are at the supply chain of the bidder will decrease under the impact of the M&A events. However, the estimated coefficients are not so significantly different from 0.

For the bidder’s customers within the horizontal sub-sample, the estimated means of the cumulative abnormal return on intervals of [-1, 1], [-2, 2] and [-5, 5] are all negative and equal to -0.11%, -0.06%, -0.53% respectively. The cumulative abnormal return on intervals of [-5, 5] is significantly different from 0 at a significance level of 10%.

Considering the bidder’s suppliers within the horizontal sub-sample, not only the estimated mean of the cumulative abnormal return on intervals of [-1, 1], [-2, 2] and [-5, 5] are all negative, but also all of them are significantly different from 0 at a significance level of 5%. The estimated means are -0.50%, -0.63% and -0.87% respectively. It is also an interesting discovery that the mean cumulative abnormal return of the bidders’ suppliers is more significantly different from zero compared with that of the bidders’ customers under the impact of the horizontal M&A events.

Regarding the difference between the means of the cumulative abnormal return of the bidders’ customers and suppliers within the horizontal sub-sample and those within the non-horizontal sub-sample, the estimated differences are all negative, which implicates that the means within the non-horizontal sub-sample are larger. The customers’ differences are not significantly different from zero while the suppliers’ differences are significantly different from zero.

Table1:Abnormal returns to related firms

    Obs N of deals Mean(%) Robust t-statistics
Full sample CAR[-1,1] of customer firms 16239 1643 -0.02163 -0.46
CAR[-2,2] of customer firms 16239 1643 -0.05338 -0.85
CAR[-5,5] of customer firms 16239 1643 -0.09294 -1.05
CAR[-1,1] of supplier firms 20580 1667 -0.05144 -0.90
CAR[-2,2] of supplier firms 20580 1667 -0.07106 -0.95
CAR[-5,5] of supplier firms 20580 1667 -0.07466 -0.66
    Obs N of deals Mean(%) Robust t-statistics Difference with non-horizontal subsample(%) Robust t-statistics
Horizontal Sub-sample CAR[-1,1] of customer firms 1265 134 -0.10693 -0.63 -0.09253 -0.53
CAR[-2,2] of customer firms 1265 134 -0.05551 -0.22 -0.00231 -0.01
CAR[-5,5] of customer firms 1265 134 -0.53188* -1.76 -0.47611 -1.51
CAR[-1,1] of supplier firms 1855 154 -0.50393 ** -2.19 -0.49732** -2.10
CAR[-2,2] of supplier firms 1855 154 -0.62843** -2.08 -0.61259** -1.97
CAR[-5,5] of supplier firms 1855 154 -0.8716 ** -2.26 -0.87591*** -2.18

***p<0.01,**p<0.05,*p<0.1

Standard errors clustered at deal level

Figure 1: Market Reaction to M&A Announcements Rating Categories

Panel A: Overall M&A

Panel B: Horizontal M&A

3. Theory Explanation for the Results

3.1 Customer          

Depending on the circumstances, the merger may be good for or bad for the customer. The reduction of competitors in the market means that enterprises may charge more for products or services. However, if the merger reduces some of the costs of previous operations, it can increase the sales volume by reducing the price, so as to improve the gross profit of the company. The combination of these two points can explain that the influence of customers in the process of merger is generally not significant enough.           

There is a significant value in the horizontal sub-sample. This may be due to the merger in the same industry or the nature of the company, which leads to a stronger monopoly position of a company in the market so that the price rises and customers are affected.

3.2 Supplier     

In a horizontal sub-sample, the change of employees, such as the change of marketing manager and advertising agent of the merging company, may have a negative impact on suppliers. The possible reason is that the general replacement of new managers may lead to the cancellation or shelving of previous supply orders, resulting in the reduction of supplier orders.     

If the two companies use the same supplier, the combined company may cut the price of the supplier because of its huge order quantity. As a result, suppliers supply the same amount of raw materials under the same conditions but sell them at a lower price. So, the negative impact is more significant in the same industry. 

In the full sample, there may be a lot of cross-industry mergers, which do not have too much interference in the original raw material orders, so there is no significant impact. 

*task 1
use MA_events,clear
describe

//1
//周末和节假日
use ret_crsp,clear
duplicates drop date,force
keep date
sort date
gen n=_n
save trading_day,replace

use MA_events,clear
gen date=DateAnnounced
joinby date using trading_day,unmatched(master)
forvalues t=1/4{
replace date=date+1 if _merge==1
drop _merge
joinby date using trading_day,unmatched(master)
}
rename date dateannounced
rename n n_announced

drop _merge
save MA_events,replace

gen cusip= AcquirorCUSIP
gen year=year(DateAnnounced)
joinby cusip year using naics
rename naics AcquirorNACIS
drop cusip
gen cusip= TargetCUSIP
joinby cusip year using naics
rename naics TargetNACIS
drop cusip
keep if AcquirorNACIS==TargetNACIS
save HorMA_events,replace

//2
use MA_events,clear
gen s_cusip= AcquirorCUSIP
gen year=year(DateAnnounced)
gen month=month(DateAnnounced)
joinby s_cusip year month using supply_chain
rename c_cusip customer
drop s_cusip year month
save customer,replace

use MA_events,clear
gen c_cusip= AcquirorCUSIP
gen year=year(DateAnnounced)
gen month=month(DateAnnounced)
joinby c_cusip year month using supply_chain
rename s_cusip supplier
drop c_cusip year month
save supplier,replace

use HorMA_events,clear
gen s_cusip= AcquirorCUSIP
gen month=month(DateAnnounced)
joinby s_cusip year month using supply_chain
rename c_cusip customer
drop s_cusip year month AcquirorNACIS TargetNACIS
save HorCustomer,replace

use HorMA_events,clear
gen c_cusip= AcquirorCUSIP
gen month=month(DateAnnounced)
joinby c_cusip year month using supply_chain
rename s_cusip supplier
drop c_cusip year month AcquirorNACIS TargetNACIS
save HorSupplier,replace

//3
use ret_crsp,clear
joinby date using ff_3factors, unmatched(master)
drop _merge
joinby date using trading_day,unmatched(master)
rename n n_date
drop _merge
save ret_ff,replace

*horcus
//数据集合并
use ret_ff,clear
rename cusip customer
joinby customer using HorCustomer
save ret_ff_horcus,replace

//分组,界定reg window和predict window
use ret_ff_horcus,clear

gen interval=n_announced-n_date
keep if interval>=-5 & interval<=100 drop if interval>5 & interval<10 egen n_event=group(AcquirorCUSIP TargetCUSIP DateAnnounced) egen n_customer=group(n_event customer) sort n_customer interval save ret_ff_horcus2,replace gen window=. replace window=1 if interval>=10 & interval<=100 replace window=2 if interval>=-5 & interval<=5 //分组回归 gen dif=ret-rf statsby, by(n_customer) saving(b_horcus):regress dif mktrf smb hml if window==1 //预测 keep if window==2 joinby n_customer using b_horcus,unmatched(master) drop _merge gen ar=. replace ar=ret-( _b_cons+ _b_mktrf*mktrf+ _b_smb*smb+ _b_hml* hml+ rf) save ar_horcus,replace *horsup use ret_ff,clear rename cusip supplier joinby supplier using HorSupplier save ret_ff_horsup,replace //分组,界定reg window和predict window use ret_ff_horsup,clear gen interval=n_announced-n_date keep if interval>=-5 & interval<=100 drop if interval>5 & interval<10 egen n_event=group(AcquirorCUSIP TargetCUSIP DateAnnounced) egen n_supplier=group(n_event supplier) sort n_supplier interval save ret_ff_horsup2,replace gen window=. replace window=1 if interval>=10 & interval<=100 replace window=2 if interval>=-5 & interval<=5 //分组回归 gen dif=ret-rf statsby, by(n_supplier) saving(b_horsup):regress dif mktrf smb hml if window==1 //预测 keep if window==2 joinby n_supplier using b_horsup,unmatched(master) drop _merge gen ar=. replace ar=ret-( _b_cons+ _b_mktrf*mktrf+ _b_smb*smb+ _b_hml* hml+ rf) save ar_horsup,replace *cus use ret_ff,clear rename cusip customer joinby customer using customer save ret_ff_cus,replace //分组,界定reg window和predict window use ret_ff_cus,clear gen interval=n_announced-n_date keep if interval>=-5 & interval<=100 drop if interval>5 & interval<10 egen n_event=group(AcquirorCUSIP TargetCUSIP DateAnnounced) egen n_customer=group(n_event customer) sort n_customer interval save ret_ff_cus2,replace gen window=. replace window=1 if interval>=10 & interval<=100 replace window=2 if interval>=-5 & interval<=5 //分组回归 gen dif=ret-rf statsby, by(n_customer) saving(b_cus):regress dif mktrf smb hml if window==1 //预测 keep if window==2 joinby n_customer using b_cus,unmatched(master) drop _merge gen ar=. replace ar=ret-( _b_cons+ _b_mktrf*mktrf+ _b_smb*smb+ _b_hml* hml+ rf) save ar_cus,replace *sup use ret_ff,clear rename cusip supplier joinby supplier using supplier save ret_ff_sup,replace //分组,界定reg window和predict window use ret_ff_sup,clear gen interval=n_announced-n_date keep if interval>=-5 & interval<=100 drop if interval>5 & interval<10 egen n_event=group(AcquirorCUSIP TargetCUSIP DateAnnounced) egen n_supplier=group(n_event supplier) sort n_supplier interval save ret_ff_sup2,replace gen window=. replace window=1 if interval>=10 & interval<=100 replace window=2 if interval>=-5 & interval<=5 //分组回归 gen dif=ret-rf statsby, by(n_supplier) saving(b_sup):regress dif mktrf smb hml if window==1 //预测 keep if window==2 joinby n_supplier using b_sup,unmatched(master) drop _merge gen ar=. replace ar=ret-( _b_cons+ _b_mktrf*mktrf+ _b_smb*smb+ _b_hml* hml+ rf) save ar_sup,replace //4 use ar_horcus,clear gen day=-interval save ar_horcus,replace bys n_customer:egen car1=total(ar) if day>=-1&day<=1 bys n_customer:egen car2=total(ar) if day>=-2&day<=2 bys n_customer:egen car5=total(ar) if day>=-5&day<=5 keep if day==0 save car_horcus,replace use car_horcus,clear reg car1, vce(cluster n_event) reg car2, vce(cluster n_event) reg car5, vce(cluster n_event) use ar_horsup,clear gen day=-interval save ar_horsup,replace bys n_supplier:egen car1=total(ar) if interval>=-1&interval<=1 bys n_supplier:egen car2=total(ar) if interval>=-2&interval<=2 bys n_supplier:egen car5=total(ar) if interval>=-5&interval<=5 keep if interval==0 save car_horsup,replace use car_horsup,clear reg car1, vce(cluster n_event) reg car2, vce(cluster n_event) reg car5, vce(cluster n_event) use ar_cus,clear gen day=-interval save ar_cus,replace bys n_customer:egen car1=total(ar) if interval>=-1&interval<=1 bys n_customer:egen car2=total(ar) if interval>=-2&interval<=2 bys n_customer:egen car5=total(ar) if interval>=-5&interval<=5 keep if interval==0 save car_cus,replace use car_cus,clear reg car1, vce(cluster n_event) reg car2, vce(cluster n_event) reg car5, vce(cluster n_event) reg car1, vce(robust) reg car2, vce(robust) reg car5, vce(robust) use ar_sup,clear gen day=-interval save ar_sup,replace bys n_supplier:egen car1=total(ar) if interval>=-1&interval<=1 bys n_supplier:egen car2=total(ar) if interval>=-2&interval<=2 bys n_supplier:egen car5=total(ar) if interval>=-5&interval<=5 keep if interval==0 save car_sup,replace use car_sup,clear reg car1, vce(cluster n_event) reg car2, vce(cluster n_event) reg car5, vce(cluster n_event) reg car1, vce(robust) reg car2, vce(robust) reg car5, vce(robust) //检验diff use car_cus,clear joinby AcquirorCUSIP TargetCUSIP DateAnnounced customer using car_horcus,unmatched(master) gen dummy=. replace dummy=1 if _merge==3 replace dummy=0 if _merge!=3 reg car1 dummy,vce(cluster n_event) reg car2 dummy,vce(cluster n_event) reg car5 dummy,vce(cluster n_event) use car_sup,clear joinby AcquirorCUSIP TargetCUSIP DateAnnounced supplier using car_horsup,unmatched(master) gen dummy=. replace dummy=1 if _merge==3 replace dummy=0 if _merge!=3 reg car1 dummy,vce(cluster n_event) reg car2 dummy,vce(cluster n_event) reg car5 dummy,vce(cluster n_event) //作图 use ar_horcus,clear mat ar_horcus=(.,.,.,.) forvalues i=-5/5{ reg ar if day==i’,vce(cluster n_event)
mat b=e(b)
mat v=e(V)
sca mean=100*b[1,1]
sca se=100*sqrt(v[1,1])
sca t=mean/se
sca low=mean-1.96*se
sca high=mean+1.96*se
mat ar_horcus=(ar_horcus\(mean,t,low,high))
}
svmat ar_horcus
keep ar_horcus*
keep in 2/12
ren ar_horcus1 AR
ren ar_horcus2 t
ren ar_horcus3 low
ren ar_horcus4 high
gen day=_n-6
twoway(connected AR day,sort)(rcap low high day,sort lpattern(dash_3dot)),ytitle(AR) yline(0,lcolor(red))xtitle(day)xlabel(#11)title(Effect of horizontal M&A deals on the portfolio of bidder’s customers,size(medium) margin(medium)) legend(off)
graph export horcus.png,replace

use ar_horsup,clear
mat ar_horsup=(.,.,.,.)
forvalues i=-5/5{
reg ar if day==
i’,vce(cluster n_event)
mat b=e(b)
mat v=e(V)
sca mean=100*b[1,1]
sca se=100*sqrt(v[1,1])
sca t=mean/se
sca low=mean-1.96*se
sca high=mean+1.96*se
mat ar_horsup=(ar_horsup\(mean,t,low,high))
}
svmat ar_horsup
keep ar_horsup*
keep in 2/12
ren ar_horsup1 AR
ren ar_horsup2 t
ren ar_horsup3 low
ren ar_horsup4 high
gen day=_n-6
twoway(connected AR day,sort)(rcap low high day,sort lpattern(dash_3dot)),ytitle(AR) yline(0,lcolor(red))xtitle(day)xlabel(#11)title(Effect of horizontal M&A deals on the portfolio of bidder’s suppliers,size(medium) margin(medium)) legend(off)
graph export horsup.png,replace

use ar_cus,clear
mat ar_cus=(.,.,.,.)
forvalues i=-5/5{
reg ar if day==i',vce(cluster n_event)
mat b=e(b)
mat v=e(V)
sca mean=100*b[1,1]
sca se=100*sqrt(v[1,1])
sca t=mean/se
sca low=mean-1.96*se
sca high=mean+1.96*se
mat ar_cus=(ar_cus\(mean,t,low,high))
}
svmat ar_cus
keep ar_cus*
keep in 2/12
ren ar_cus1 AR
ren ar_cus2 t
ren ar_cus3 low
ren ar_cus4 high
gen day=_n-6
twoway(connected AR day,sort)(rcap low high day,sort lpattern(dash_3dot)),ytitle(AR) yline(0,lcolor(red))xtitle(day)xlabel(#11)title(Effect of overall M&A deals on the portfolio of bidder's customers,size(medium) margin(medium)) legend(off)
graph export cus.png,replace

use ar_sup,clear
mat ar_sup=(.,.,.,.)
forvalues i=-5/5{
reg ar if day==
i’,vce(cluster n_event)
mat b=e(b)
mat v=e(V)
sca mean=100*b[1,1]
sca se=100*sqrt(v[1,1])
sca t=mean/se
sca low=mean-1.96*se
sca high=mean+1.96*se
mat ar_sup=(ar_sup\(mean,t,low,high))
}
svmat ar_sup
keep ar_sup*
keep in 2/12
ren ar_sup1 AR
ren ar_sup2 t
ren ar_sup3 low
ren ar_sup4 high
gen day=_n-6
twoway(connected AR day,sort)(rcap low high day,sort lpattern(dash_3dot)),ytitle(AR) yline(0,lcolor(red))xtitle(day)xlabel(#11)title(Effect of overall M&A deals on the portfolio of bidder’s suppliers,size(medium) margin(medium)) legend(off)
graph export sup.png,replace

分类: Polyu

elijahqi

退役了 现在在商院 偶尔打CF,有时有ACM regional也去玩一下

发表评论