You can do this with the lag() function from dplyr.
df <- data.frame( y = c(1137, 1159))
library(dplyr)
#>
#> Attaching package: 'dplyr'
#> The following objects are masked from 'package:stats':
#>
#> filter, lag
#> The following objects are masked from 'package:base':
#>
#> intersect, setdiff, setequal, union
df <- df %>% mutate(lagValue = lag(y), Perc = (y/lagValue) - 1)
df
#> y lagValue Perc
#> 1 1137 NA NA
#> 2 1159 1137 0.01934916