The table below shows the association for the number of Airbnb listings and Calls for Service in the same month. There is a positive correlation, as Airbnb’s are located in areas that have more calls for service across all six types of calls.
A positive varaible indicates a postive correaltion for the data, while a negative sign indicates the opposite. Stars(*) indicate how confident we are in the results, with three stars indicating we’re pretty confident.
Lots of intersting thigns to unpack here.
##
## ================================================================================================
## Dependent variable:
## ---------------------------------------------------------------
## Number of Calls for Service
## Phys Prop Pers Prop Soc Dis Phys Dis Violent Sex
## (1) (2) (3) (4) (5) (6)
## ------------------------------------------------------------------------------------------------
## Number of Airbnbs (log) 0.266*** 0.045*** 0.291*** 0.151*** 0.110*** 0.004***
## (0.007) (0.003) (0.006) (0.006) (0.005) (0.001)
##
## % White -0.302*** -0.047 -0.440*** -0.169*** -0.228*** 0.039***
## (0.066) (0.032) (0.062) (0.065) (0.046) (0.014)
##
## % Black 0.126* -0.033 0.229*** 0.045 0.165*** 0.089***
## (0.069) (0.033) (0.064) (0.067) (0.048) (0.015)
##
## % Housing Vacant 0.559*** 0.165*** 0.815*** 0.505*** 0.693*** 0.072***
## (0.063) (0.030) (0.059) (0.062) (0.044) (0.013)
##
## % College Grduate 0.797*** 0.178*** 0.306*** 0.399*** -0.090*** -0.055***
## (0.045) (0.022) (0.042) (0.044) (0.032) (0.010)
##
## Median Income (log) -0.384*** -0.120*** -0.605*** -0.290*** -0.425*** 0.008**
## (0.018) (0.009) (0.017) (0.018) (0.012) (0.004)
##
## Median Rent (log) 0.098*** 0.026*** 0.041** 0.053** 0.005 0.010**
## (0.021) (0.010) (0.020) (0.021) (0.015) (0.005)
##
## Population (log) 0.286*** 0.100*** 0.367*** 0.289*** 0.295*** 0.021***
## (0.012) (0.006) (0.011) (0.012) (0.008) (0.003)
##
## Constant 1.693*** 0.371*** 5.035*** 1.297*** 3.153*** -0.311***
## (0.252) (0.120) (0.233) (0.246) (0.175) (0.053)
##
## ------------------------------------------------------------------------------------------------
## Neighbourhood FE Yes Yes Yes Yes Yes Yes
## Time FE Yes Yes Yes Yes Yes Yes
## Observations 42,101 42,101 42,101 42,101 42,101 42,101
## R2 0.434 0.294 0.511 0.359 0.427 0.166
## Adjusted R2 0.431 0.290 0.508 0.356 0.424 0.161
## Residual Std. Error (df = 41866) 0.861 0.409 0.798 0.841 0.597 0.183
## F Statistic (df = 234; 41866) 137.080*** 74.382*** 187.076*** 100.318*** 133.422*** 35.491***
## ================================================================================================
## Note: *p<0.1; **p<0.05; ***p<0.01
##
## ===============================
## Three Cities - Same Time Period
## -------------------------------The table below reports how changes in the previous month (August to September) impact changes in the calls for service (from September to October). The on month difference ensures that change sin Airbnb preceed the changes in calls for service, stregthening the causal argument.
As Airbnb’s increase in a block, there is an increase in the calls for service for personal property and social disorder. However, there is a decrease in calls for service realting to violent acts.
stargazer(fep11, fep12, fep13, fep14, fep15, fep16, type="text", dep.var.labels ="Change in Calls for Service",
main="Three Cities - Changes",
column.labels = c("Phys Prop", "Pers Prop", "Soc Dis", "Phys Dis", "Violent", "Sex"),
covariate.labels = c("Change in Airbnb from Previous Month (log)",
"% White",
"% Black",
"% Housing Vacant",
"% College Grduate",
"Median Income (log)",
"Median Rent (log)",
"Population (log)"), omit=c("neighbourhood", "order"),
add.lines=list(c("Neighbourhood FE", "Yes", "Yes", "Yes", "Yes", "Yes", "Yes"), c("Time FE", "Yes", "Yes", "Yes", "Yes", "Yes", "Yes")))##
## ===================================================================================================
## Dependent variable:
## --------------------------------------------------------
## Change in Calls for Service
## Phys Prop Pers Prop Soc Dis Phys Dis Violent Sex
## (1) (2) (3) (4) (5) (6)
## ---------------------------------------------------------------------------------------------------
## Change in Airbnb from Previous Month (log) -0.006 0.024*** 0.012* 0.011 -0.029*** -0.004
## (0.007) (0.005) (0.007) (0.009) (0.009) (0.008)
##
## % White -0.004 0.004 -0.003 -0.001 0.0003 -0.003
## (0.004) (0.003) (0.004) (0.006) (0.005) (0.005)
##
## % Black -0.0001 0.002 -0.001 0.003 0.002 -0.003
## (0.004) (0.003) (0.004) (0.006) (0.005) (0.005)
##
## % Housing Vacant -0.006 0.004 -0.004 -0.006 -0.003 -0.005
## (0.004) (0.003) (0.004) (0.005) (0.005) (0.004)
##
## % College Grduate 0.005* -0.001 0.005* 0.002 0.001 0.003
## (0.003) (0.002) (0.003) (0.004) (0.004) (0.003)
##
## Median Income (log) -0.001 0.0004 -0.001 0.001 0.001 -0.001
## (0.001) (0.001) (0.001) (0.001) (0.001) (0.001)
##
## Median Rent (log) -0.003** -0.001 -0.002* 0.00003 -0.001 0.00002
## (0.001) (0.001) (0.001) (0.002) (0.002) (0.002)
##
## Population (log) -0.002** -0.001 -0.001 -0.0003 -0.001 -0.001
## (0.001) (0.001) (0.001) (0.001) (0.001) (0.001)
##
## Constant 6.576*** 4.083*** 6.126*** 4.892*** 4.012*** 2.530***
## (0.041) (0.029) (0.037) (0.053) (0.051) (0.045)
##
## ---------------------------------------------------------------------------------------------------
## Neighbourhood FE Yes Yes Yes Yes Yes Yes
## Time FE Yes Yes Yes Yes Yes Yes
## Observations 29,707 29,707 29,707 29,707 29,707 29,707
## R2 0.030 0.010 0.023 0.010 0.012 0.019
## Adjusted R2 0.022 0.002 0.015 0.003 0.004 0.012
## Residual Std. Error (df = 29485) 0.046 0.033 0.042 0.059 0.057 0.050
## F Statistic (df = 221; 29485) 4.070*** 1.288*** 3.092*** 1.390*** 1.607*** 2.584***
## ===================================================================================================
## Note: *p<0.1; **p<0.05; ***p<0.01
##
## ======================
## Three Cities - Changes
## ----------------------We can look at New Orleans seperately to see if these trends are just driven by tourism, or whether Airbnb is having an independent effect. For New Orleans we have data on wherhe hotels are located, which we can use to proxy for typical tourism.
In the first table below, we see that Airbnb’s locate in areas with more calls for service in New orleans. The results are generally similar.
##
## ================================================================================================
## Dependent variable:
## ---------------------------------------------------------------
## Number of Calls for Service
## Phys Prop Pers Prop Soc Dis Phys Dis Violent Sex
## (1) (2) (3) (4) (5) (6)
## ------------------------------------------------------------------------------------------------
## Number of Airbnbs (log) 0.210*** 0.052*** 0.231*** 0.074*** 0.061*** 0.011***
## (0.008) (0.006) (0.007) (0.010) (0.007) (0.003)
##
## Number of Hotels (log) 0.213*** 0.168*** 0.337*** 0.358*** 0.233*** 0.032***
## (0.017) (0.012) (0.014) (0.019) (0.013) (0.007)
##
## % White 0.757*** 0.157** 0.486*** 0.698*** 0.375*** 0.051
## (0.114) (0.080) (0.098) (0.133) (0.090) (0.046)
##
## % Black 1.714*** 0.402*** 1.825*** 1.307*** 1.072*** 0.162***
## (0.105) (0.073) (0.091) (0.123) (0.083) (0.042)
##
## % Housing Vacant 0.457*** 0.197*** 0.919*** 0.317*** 0.773*** 0.124***
## (0.068) (0.048) (0.059) (0.080) (0.054) (0.027)
##
## % College Grduate 1.582*** 0.350*** 0.910*** 0.644*** 0.223*** -0.123***
## (0.059) (0.041) (0.051) (0.069) (0.047) (0.024)
##
## Median Income (log) -0.184*** -0.062*** -0.456*** -0.120*** -0.327*** 0.022**
## (0.022) (0.015) (0.019) (0.026) (0.017) (0.009)
##
## Median Rent (log) 0.161*** 0.001 0.003 0.011 -0.044* 0.024**
## (0.029) (0.020) (0.025) (0.033) (0.023) (0.011)
##
## Population (log) 0.371*** 0.173*** 0.516*** 0.293*** 0.366*** 0.056***
## (0.017) (0.012) (0.014) (0.019) (0.013) (0.007)
##
## Constant -1.818*** -0.839*** 1.305*** -2.149*** 0.973*** -0.854***
## (0.270) (0.189) (0.234) (0.316) (0.214) (0.109)
##
## ------------------------------------------------------------------------------------------------
## Neighbourhood FE Yes Yes Yes Yes Yes Yes
## Time FE Yes Yes Yes Yes Yes Yes
## Observations 16,920 16,920 16,920 16,920 16,920 16,920
## R2 0.432 0.296 0.620 0.479 0.490 0.160
## Adjusted R2 0.428 0.291 0.618 0.475 0.487 0.155
## Residual Std. Error (df = 16806) 0.695 0.486 0.601 0.812 0.551 0.279
## F Statistic (df = 113; 16806) 113.166*** 62.483*** 243.144*** 136.598*** 142.867*** 28.411***
## ================================================================================================
## Note: *p<0.1; **p<0.05; ***p<0.01
##
## ==============================
## New Orleans - Same Time Period
## ------------------------------The table below shows how changes in the previous month impact changes in calls for service in New Orleans. We have more confidence in these results, with social disorder and personal property both increaseing. Violence realted calls still decreases, and there is an odd increase in physical property related calls.
stargazer(fep11, fep12, fep13, fep14, fep15, fep16, type="text", dep.var.labels ="Number of Calls for Service",
main="New Orleans - Changes",
column.labels = c("Phys Prop", "Pers Prop", "Soc Dis", "Phys Dis", "Violent", "Sex"),
covariate.labels = c("Change in Airbnb in Previous Month (log)",
"Number of Hotels (log)",
"% White",
"% Black",
"% Housing Vacant",
"% College Grduate",
"Median Income (log)",
"Median Rent (log)",
"Population (log)"), omit=c("neighbourhood", "order"),
add.lines=list(c("Neighbourhood FE", "Yes", "Yes", "Yes", "Yes", "Yes", "Yes"),
c("Time FE", "Yes", "Yes", "Yes", "Yes", "Yes", "Yes")))##
## =================================================================================================
## Dependent variable:
## --------------------------------------------------------
## Number of Calls for Service
## Phys Prop Pers Prop Soc Dis Phys Dis Violent Sex
## (1) (2) (3) (4) (5) (6)
## -------------------------------------------------------------------------------------------------
## Change in Airbnb in Previous Month (log) 0.048*** 0.033*** 0.326*** 0.011 -0.106*** 0.002
## (0.008) (0.008) (0.017) (0.011) (0.021) (0.013)
##
## Number of Hotels (log) -0.001 -0.003** -0.005** -0.002 -0.006** -0.003
## (0.001) (0.001) (0.002) (0.002) (0.003) (0.002)
##
## % White -0.004 0.016* 0.003 0.002 0.014 -0.009
## (0.009) (0.008) (0.017) (0.011) (0.021) (0.013)
##
## % Black -0.003 0.009 -0.002 0.001 0.005 -0.009
## (0.008) (0.008) (0.015) (0.010) (0.020) (0.012)
##
## % Housing Vacant -0.002 0.007 -0.007 0.002 -0.001 -0.007
## (0.005) (0.005) (0.010) (0.007) (0.013) (0.008)
##
## % College Grduate -0.0003 -0.001 -0.003 0.007 -0.003 0.006
## (0.004) (0.004) (0.008) (0.006) (0.011) (0.007)
##
## Median Income (log) 0.001 0.00005 0.002 -0.001 0.005 -0.002
## (0.002) (0.002) (0.003) (0.002) (0.004) (0.003)
##
## Median Rent (log) -0.002 -0.001 -0.001 0.001 -0.003 -0.0003
## (0.002) (0.002) (0.004) (0.003) (0.005) (0.003)
##
## Population (log) -0.001 -0.001 -0.001 0.001 -0.005* -0.002
## (0.001) (0.001) (0.002) (0.002) (0.003) (0.002)
##
## Constant 4.616*** 4.035*** 2.630*** 4.896*** 3.645*** 2.520***
## (0.049) (0.047) (0.096) (0.064) (0.122) (0.075)
##
## -------------------------------------------------------------------------------------------------
## Neighbourhood FE Yes Yes Yes Yes Yes Yes
## Time FE Yes Yes Yes Yes Yes Yes
## Observations 13,959 13,959 13,959 13,959 13,959 13,959
## R2 0.025 0.013 0.051 0.017 0.011 0.035
## Adjusted R2 0.017 0.005 0.044 0.010 0.003 0.028
## Residual Std. Error (df = 13852) 0.048 0.046 0.093 0.063 0.119 0.073
## F Statistic (df = 106; 13852) 3.338*** 1.706*** 7.036*** 2.313*** 1.427*** 4.755***
## =================================================================================================
## Note: *p<0.1; **p<0.05; ***p<0.01
##
## =====================
## New Orleans - Changes
## ---------------------