t-value and p-value seem wrong?

I have a data-frame. Downloaded from http://www.nyc.gov/html/tlc/html/about/trip_record_data.shtml. My dataset is from 2018 and the month of January. I keep these columns: trip_distance, fare_amount, pickup_time and dropoff_time.

enter image description here

The goal is to calculate 'price_per_mile'. Then, the mean of these values for each borough and then, applying the t-test to see if the differences among each pair of them are significant. The problem is that at the end I get t-values=0 and p-values=1 for all the pairs (just one exception). I don't understand what are the things I need to recheck or change? You can reach 'taxi_zone_lookup.csv' from this address too: http://www.nyc.gov/html/tlc/html/about/trip_record_data.shtml

this is my code:

df=pd.read_csv('yellow_tripdata_2018-01.csv', 
             usecols=['tpep_pickup_datetime', 'tpep_dropoff_datetime','trip_distance','PULocationID','fare_amount'])


#Data cleaning 
df.drop(df[df['trip_distance']>3].index, inplace=True)
df.drop(df[df['trip_distance']<0.5].index, inplace=True) 

df.drop(df[df['fare_amount']>10].index, inplace=True)
df.drop(df[df['fare_amount']<1].index, inplace=True) 

df['trip_distance']=df['trip_distance'].astype(np.float16)
df['PULocationID']=df['PULocationID'].astype(np.uint16)
df['fare_amount']=df['fare_amount'].astype(np.float16)

df['price_per_mile'] = df['fare_amount']/df['trip_distance']

borough = pd.read_csv(r'taxi_zone_lookup.csv', usecols = ['LocationID', 'Borough'])

result = pd.merge(df,
             borough,
             left_on='PULocationID',
             right_on='LocationID',
             how='inner' 
             )

result.drop(result[(result.Borough == 'EWR') | (result.Borough == 'Unknown')].index, inplace=True)

df['price_per_mile'].describe()
#here I get mean=NaN???

#t-test
#Creating a data-frame with two-level of indexes
boroughs = ['Bronx', 'Brooklyn', 'Manhattan', 'Staten Island', 'Queens']
iterables = [['Bronx', 'Brooklyn', 'Manhattan', 'Staten Island', 'Queens'], ['t-value', 'p-value', "H0 hypothesis"]]
my_index = pd.MultiIndex.from_product(iterables)
dt = pd.DataFrame(index=my_index, columns=boroughs)

for i in boroughs:
  a = result.loc[result.Borough==i]["price_per_mile"]
     for j in boroughs:
         b = result.loc[result.Borough==j]["price_per_mile"]
         t2, p2 = stats.ttest_ind(a,b)
         dt.loc[(i,"t-value"),j]=t2
         dt.loc[(i,"p-value"),j]=p2
         if(p2>0.05):
            dt.loc[(i,"H0 hypothesis"),j]='Fail to Reject H0'
         else:
            dt.loc[(i,"H0 hypothesis"),j]='Reject H0'