The issues related to outbound MQLs are interesting. I see two issues that can arise if outbound MQLs don't exist.
1) You might miscalculate the conversion rate from MQLs to SQLs.
For example, in a situation like this:
Total Leads: 200 (100 outbound, 100 inbound)
Total MQLs: 80 (all from inbound, as outbound leads are not marked as MQLs)
Total SQLs: 140 (60 from inbound, 80 from outbound)
Conversion rate from MQLs to SQLs: might be miscalculated as 175% (140/80) but should be 75% (60/80)
2) You can't compare the conversion rate from MQLs to SQLs between the inbound and outbound funnel.
If you lack outbound MQLs (assuming you use lead scoring), you lack an objective measure of the quality of your outbound leads. There are no conversions from lead to MQL for outbound leads. You also lack a measure of how well the good outbound leads (outbound MQLs) convert to meetings compared to inbound MQLs. If you don't have outbound MQLs, it's hard to know if you need better leads (more MQLs) or a better prospecting process (a better MQL to SQL conversion rate).
This is a challenging situation in general.
If you do use outbound MQLs, you don't have these two issues. But, you still need to be careful when looking at the conversion rates for all leads because you lose detail. If one motion or channel converts better, it will raise the average.
Regardless of outbound MQLs, conversion rates should be segmented by channel.
Anything else I'm missing?