Evaluation of Tax Credits

This chapters discusses the inclusion of the federal programs designed to help low-income communities.

Rodeo Dataset The data in the below table includes the first 10 rows of the 2000 & 2010 census data, with the transformed variables.

The smaller dataset includes four variables that examines different facets of the median home values in 2000 & 2010, the change and growth.

The data set for difference-in-difference model uses the log transformation of the median house value, dummy variables for the treatment, counterfactual, and marking the periods before & after the federal programs.

Rodeo Data Set The entire dataset was combined into the data rodeo for easier use.

## Data Manifest The full dataset consists of sample & full data set, and their combined data sets for 2000 & 2010. Finally, the 2000 & 2010 data sets were combined into one file.

Rows in the 2000 sample & full data sets

The sample and the full data sets have an equal number of observations.

## [1] 72693

## [1] 72693

Rows in the 2000 rodeo

The combined data set for 2000 has the same number of observations as the sample and full data sets.

## [1] 72693

Rows in 2010 sample and full data sets

The observations in the 2010 data sets are not equal because the new variables were created in the full census data set.

## [1] 73056

## [1] 74002

Rows in 2010 data rodeo

The data rodeo has a larger number of variables to contain the transformed variables in the full census data set.

## [1] 74022

Rows in complete data rodeo

The observations in the final data set are fewer than the combined 2010 or 2000 data set because of data wrangling.The data wrangling included the exclusion of all rural tracts. Additionally, tracts with plots valued at less than $2,000 were dropped to ignore vacant lots. Finally, homes that experienced median home growth over 200% were excluded as outliers.

## [1] 59066

Descriptive Statistics

Recipient & Non-Recipient Tracts

## 
## ==========================================================================
## Statistic              Min     Pctl(25) Median   Mean   Pctl(75)    Max   
## --------------------------------------------------------------------------
## MedianHomeValue2000   11,167   105,661  154,903 187,129 224,337  1,288,551
## MedianHomeValue2010   9,999    123,200  193,200 246,570 312,000  1,000,001
## MHV.Change.00.to.10 -1,228,651  7,187   36,268  60,047   94,881  1,000,001
## MHV.Growth.00.to.12    -97        6       25      33       50      6,059  
## --------------------------------------------------------------------------

Participant and Non-Participant Tracts - Poverty Rates

The recipients of the New Market Tax Credits (NMTC) tended to be in areas with moderate to high levels of poverty.

The LIHTC projects concentrated more funding in tracts with lower rates of poverty but had a higher concentration in tracts with a higher density of poverty.
Home Values

The median home values of the program recipients follows a similar pattern to the poverty rates. The NMTC recipient MHV was lower than the non-recipients. While the LIHTC recipients were more densely concentrated in the lowest MHV tracts but had a higher concentration in the higher MHV than the NMTC.

Household Income 2000 - LIHTC

Another way to visualize the differences between the participation and non-participating households in each program are the average incomes of each household type.

Average income of the tractsparticipatingin the LIHTC program.

## [1] 47302.67

Average income of the non-participating tracts was 24% higher than the participating tracts.

## [1] 63789.16

Household Income 2000 - NMTC

In general, the income of the NMTC households skewed lower than the LIHTC households.

Household income of the participating tracts

## [1] 35792.88

Median household income of non-participating tracts

The household that did not participate had income of more than 50% lower than the non-participants. This confirms the participating household concentration in tracts with a higher density of poverty.

## [1] 62259.52

Median Home Value in 2000 - LIHTC

Median home value of the participating tracts in the LIHTC program

## [1] 156532.2

Median home value of non-participating tracts in the LIHTC program

The mean home values that did not participate were roughly 19% higher than the participating tracts.

## [1] 192122.5

Median Home Value in 2000 - NMTC program

Median home value of the participating households in the NMTC program

The NMTC participating households had a lower median home value than the LIHTC program.

## [1] 146691.6

Median home value of the non-participating tracts in the NMTC program

The tracts that participated had home values roughly 22% lower than the non-participating tracts.

## [1] 188383

2000 Poverty Rates - LIHTC

Poverty rate of the participating tracts in the LIHTC program

## [1] 19.33965

Poverty rate of the non-participating tracts in the LIHTC Program

The participation tracts had almost double the rate of poverty than the non-participating tracts.

## [1] 11.20438

Poverty rate of the participating tracts in the NMTC Program

Following the trend discussed above, the NMTC participation tracts had a far higher concentration of poverty than the LIHTC program.

## [1] 27.28432

Poverty rate of the non-participating tracts in the NMTC program

The poverty rate of the non-participant tracts was similar across both of the programs.

## [1] 11.89316

Median Home Value Growth Rates in the LIHTC & NMTC programs

The outcome of the programs was measured by the 2000 Median House Value Growth. The LIHTC participants had a higher MHV growth in comparison to the NMTC program. The outcome is not surprising because the program participation is more concentrated in areas with a higher rate of poverty, lower household income and lower MHV before participation.

Median House Value Growth of the participating tracts in the LIHTC program

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
## -88.967   4.216  23.006  29.821  48.740 199.739     115

Median House Value Growth of the non-participating tracts in the LHITC program

The MHV growth in households that participated in the program experienced a lower rate of growth than the non-participation.

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max.    NA's 
## -96.891   5.963  25.160  29.307  48.678 199.550     385

Median house value of the participating tracts in the NMTC Program

While the households that participated in the NMTC program experienced a higher rate of median home value growth than non-participants and in the LHITC program.

##    Min. 1st Qu.  Median    Mean 3rd Qu.    Max. 
##  -94.55   11.86   35.03   54.50   66.09 4334.67

Median house value of the non-participating tracts in the NMTC Program

The tracts that did not participate has a lower growth than the non-LHITC participants.

##     Min.  1st Qu.   Median     Mean  3rd Qu.     Max. 
##  -96.891    5.749   24.987   32.172   49.049 6059.261

Program Impact Evaluation
NMTC Regression Model Program Evaluation

Median Home Value & the percentage of residents with a college degree In the NMTC model in the comparison group without the control group had decrease in MHV of (-0.26) & did improve but not a statistically significant amount. However, the inclusion of the control variable, (the percentage of college graduate residents) the MHV increased by a significant amount (1.33).After the program, the MHV increased (0.14) but the houses that received the treatment did not have a statistically significant increase (0.02).

## 
## ============================================================================================================
##                                                       Dependent variable:                                   
##                     ----------------------------------------------------------------------------------------
##                                                        Median Home Value                                    
##                                  (1)                          (2)                           (3)             
## ------------------------------------------------------------------------------------------------------------
##                               10.24***                      11.96***                     10.19***           
##                                (0.01)                       (0.003)                       (0.01)            
##                                                                                                             
## p.col                          1.35***                                                    1.33***           
##                                (0.005)                                                    (0.005)           
##                                                                                                             
## treat                                                       -0.26***                       0.01             
##                                                              (0.02)                       (0.01)            
##                                                                                                             
## post                                                        0.23***                       0.14***           
##                                                             (0.004)                       (0.003)           
##                                                                                                             
## treat:post                                                  0.10***                        0.02             
##                                                              (0.02)                       (0.02)            
##                                                                                                             
## ------------------------------------------------------------------------------------------------------------
## Observations                   118,132                      118,132                       118,132           
## R2                              0.43                          0.04                         0.44             
## Adjusted R2                     0.43                          0.04                         0.44             
## Residual Std. Error      0.49 (df = 118130)            0.64 (df = 118128)           0.49 (df = 118127)      
## F Statistic         87,400.24*** (df = 1; 118130) 1,432.34*** (df = 3; 118128) 22,985.08*** (df = 4; 118127)
## ============================================================================================================
## Note:                                                                            *p<0.1; **p<0.05; ***p<0.01

LIHTC Regression Model Program Evaluation

Median Home Value & the percentage of residents with a college degree The LIHTC model without the control variable (% of college graduate residents) the median home value (MHV) the tracts participating in the program is lower at -0.08 versus tracts that did not receive treatment 0.23. The tracts that did receive the treatment had a larger, but statistically insignificant amount of 0.02. However, the inclusion of the control variable (percent of residents with a college education) had a higher MHV of 1.33, however the tracts that received the treatment had a higher MHV of 0.09. Additionally, the tracts that did not participate in the program had a lower MHV growth of 0.14 compared to the model without the control variable. Finally, the tracts that did participate in the program had a much smaller and insignificant MHV increase of 0.004.

## 
## ============================================================================================================
##                                                       Dependent variable:                                   
##                     ----------------------------------------------------------------------------------------
##                                                        Median Home Value                                    
##                                  (1)                          (2)                           (3)             
## ------------------------------------------------------------------------------------------------------------
## Constant                      10.24***                      11.98***                     10.20***           
##                                (0.01)                       (0.003)                       (0.01)            
##                                                                                                             
## % college grads                1.35***                                                    1.32***           
##                                (0.005)                                                    (0.005)           
##                                                                                                             
## treatment                                                   -0.21***                     -0.02***           
##                                                              (0.01)                       (0.01)            
##                                                                                                             
## no treatment                                                0.23***                       0.15***           
##                                                             (0.004)                       (0.003)           
##                                                                                                             
## post treatment                                                0.01                         -0.01            
##                                                              (0.01)                       (0.01)            
##                                                                                                             
## ------------------------------------------------------------------------------------------------------------
## Observations                   118,132                      118,132                       118,132           
## R2                              0.43                          0.04                         0.44             
## Adjusted R2                     0.43                          0.04                         0.44             
## Residual Std. Error      0.49 (df = 118130)            0.64 (df = 118128)           0.49 (df = 118127)      
## F Statistic         87,400.24*** (df = 1; 118130) 1,799.55*** (df = 3; 118128) 23,000.68*** (df = 4; 118127)
## ============================================================================================================
## Note:                                                                            *p<0.1; **p<0.05; ***p<0.01
## 
## ========================
## LIHTC Program Evaluation
## ------------------------