Time and space complexity is not determined by the language. It is the algorithm/data structure/implementation that you choose to use to solve a certain problem. C++ is fast because it provides fast execution as it does not employ runtime checks/out-of-bounds-checks, etc. unlike Java (let's say). It believes that the programmer is always correct. Hence, C++ users often end up with the most runtime errors. Similarly, C++ users need to often use modular arithmetic implementations to solve Big Integer problems, which is not needed in languages like Python, C#, Java, etc. Some people like to go really object-oriented in everything to do, hence Ruby. Some like coding in Assembly.

Every language has its pros and cons. It's the choice of the programmer to select their language. You can't enforce C++ on everyone. Hence, the diversity.