According to the Colby Chemistry Database, the bond angle of $\ce{PBr3}$ and $\ce{PCl3}$ are around $101$ degrees, but the bond angle of $\ce{PH3}$ is $92$ degrees. (Source: PCl3 PBr3 PH3)
There was a previous question on Stack Exchange about the bond angle difference in $\ce{PF3}$ and $\ce{PH3}$ here: PF3 and PH3, but the accepted answer mentions back bonding as a factor to increase the bond angle in $\ce{PF3}$. I was skeptical about this answer, because in $\ce{PCl3}$ and $\ce{PBr3}$, there should be less back bonding than in $\ce{PF3}$ because bromine and chlorine are both much larger than fluorine and there will be worse orbital overlap between the halogen. Less back bonding should make the bond angle of $\ce{PCl3}$ and $\ce{PBr3}$ smaller than that of $\ce{PF3}$, if back bonding were the major factor here? In reality, that is not the case. Why not?