Can you add a bit more detail here? What language/architecture are you using? Can you provide an example/something reproducible so we can run it?

Why are negative numbers of type int32 encoded using 64-bit varints? Instead of using 32-bit varint encoding

The designers wanted the 32-bit varints to be forward compatible with their 64-bit counterparts.

From https://developers.google.com/protocol-buffers/docs/proto3#updating

int32, uint32, int64, uint64, and bool are all compatible – this means you can change a field from one of these types to another without breaking forwards- or backwards-compatibility. If a number is parsed from the wire which doesn't fit in the corresponding type, you will get the same effect as if you had cast the number to that type in C++ (for example, if a 64-bit number is read as an int32, it will be truncated to 32 bits).

and

sint32 and sint64 are compatible with each other but are not compatible with the other integer types.

thank