For China, as others have said, that is mostly for allowing a common writing form for multiple related languages.
By replacing the Traditional writing with the Simplified writing, the Chinese have already made impossible the reading of the older writings without studying separately the Traditional writing, exactly as if they would have converted to the Latin alphabet.
The reason why Korea and Vietnam were successful in replacing the Chinese writing system, while Japan cannot do the same, is that the Korean and Vietnamese languages have been less influenced by the Chinese language than the Japanese language.
The modern written form of the Japanese language is a mixed language that uses both words from the language spoken by the Japanese before the introduction of the Chinese writing system and also a much larger number of words taken from an older form of the Chinese language.
Unfortunately, the very large number of Chinese words that have been taken together with the signs correspond to a much smaller number of Japanese pronunciations, creating an extremely large number of homophones, which cause an ambiguity in writing that is extremely difficult to avoid without the using of kanji.
An alphabetic form of Japanese writing, or writing it only with hiragana/katakana, would work only if the text would include along most words some kind of classifier sign, written with some kind of kanji or emoji, which would show the semantic class of the word, like it was done in a few ancient writing systems.
To remove ambiguities, a pretty long list of such semantic classifiers would need to be devised.
To give an example for those who are not familiar with how semantic classifier signs are used in writing systems, if there were a list of semantic classifiers that included 2 signs with the meaning "human body part" and "metal tool", then one could disambiguate the Japanese word "ken" by writing the sign for "human body part" followed by the phonetic signs for "ken", to indicate that the meaning is Chinese "quan2" = native Japanese "kobushi" = English "fist".
Similarly, one would write the sign for "metal tool" followed by the phonetic signs for "ken", to indicate that the meaning is Chinese "jian4" = native Japanese "tsurugi" = English "sword".
The Japanese word "ken" has many other meanings besides these 2, each being derived from distinct Chinese words, for example prefecture from xian4, ticket from quan4, and many others.
A large number of other semantic classifiers would be needed to disambiguate all meanings, e.g. "geographic place", "document" and so on.
So it is likely that even if a system with semantic classifiers would no longer need at least 2000 kanji signs, it would still need about one hundred semantic classifiers besides a phonetic syllabary or alphabet, to be able to write the Japanese language without losing information in comparison with the kanji writing.
By replacing the Traditional writing with the Simplified writing, the Chinese have already made impossible the reading of the older writings without studying separately the Traditional writing, exactly as if they would have converted to the Latin alphabet.
The reason why Korea and Vietnam were successful in replacing the Chinese writing system, while Japan cannot do the same, is that the Korean and Vietnamese languages have been less influenced by the Chinese language than the Japanese language.
The modern written form of the Japanese language is a mixed language that uses both words from the language spoken by the Japanese before the introduction of the Chinese writing system and also a much larger number of words taken from an older form of the Chinese language.
Unfortunately, the very large number of Chinese words that have been taken together with the signs correspond to a much smaller number of Japanese pronunciations, creating an extremely large number of homophones, which cause an ambiguity in writing that is extremely difficult to avoid without the using of kanji.
An alphabetic form of Japanese writing, or writing it only with hiragana/katakana, would work only if the text would include along most words some kind of classifier sign, written with some kind of kanji or emoji, which would show the semantic class of the word, like it was done in a few ancient writing systems.
To remove ambiguities, a pretty long list of such semantic classifiers would need to be devised.