Virus severity is a lot more objective than response rate, it is really all about how quickly the virus spreads and how sick people get.  So:

What a less severe virus would look like: 

• Lower than expected mortality rate.
• Lower than expected hospitalization rate.
• Lower transmission (lower basic reproduction, R)

What a more severe virus would look like:

• Higher than expected mortality rate
• Higher than expected hospitalization rate.
• Higher transmission (higher basic reproduction, R)

Based on what I see the severity seems to be higher than expected.  From what I read the initial assessments of mortality were thought to be high due to a lack of testing.  People with less severe symptoms did not factor into the calculations as strongly.

However, the Italian and French experience seem to support the higher estimates.  The WHO initially estimated a basic reproduction value of 2.2, but an article by Prof Andrea Remuzzi and Prof Giuseppe Remuzzie in the Lancet on March 13 suggested a higher R value of 2.75 to 3.25. 

Also, the mortality rate seems to be at the higher end of the range rather than lower.  However, it disproportionality affects older people: the majority of people who have died in Italy were over 80 years old. 

More interesting than mortality, perhaps, is the hospitalization rate which this same article suggests has been steady at 10%.

In this case social distancing will be effective and flattening the curve will be critical.  The US has the fewest number of beds per capita of any “developed” country, (2.8 per thousand, compared to Italy at 3.5).

This suggests increasing rather than decreasing severity of the Covid-19 virus.  With the caveat that since it disproportionately impacts older individuals, populations that skew younger will fare better.