It's not like that in most countries in fact. The US variance between high murder parts of high murder rate cities and the murder rate for the other 95% to 97% of the population is several times greater than other countries with comparable murder rates.
The murder rate among Sweden's largest cities does not consistently vary by ~20-30 fold top to bottom. That's the gulf between New York City and Baltimore, or Honolulu and St Louis, or Detroit and Austin TX. That extreme of a variance, is very unusual for all but a few countries.
Argentina for example, which is at least somewhat similar to the US in per capita murders in a given year, in their major cities you do not see Baltimore, Detroit, Chicago type examples of extreme outliers vs the ~6 national rate (eg Buenos Aires is typically around 5 or 6, as is Cordoba). Rosario, which saw a large murder rate spike in 2012-13, was considered shocking, because the murder rate went from ~10 to ~22 over a few years. So the US national rate is lower than Argentina, while having drastically higher outliers like Baltimore and Detroit versus eg Rosario (their bad case example).
