This means that financial asset has a lower risk than the risk-free market asset
The word "risk" does not have an universal nor well-defined definition in finance. Indeed, you need to define precisely (by that I mean mathematically) what you mean by "risk" before you use the word (hence why I avoided it in my explanation). There have been seminal papers that have shown
to encapsulate all kinds of risk: systematic, systemic, idiosyncratic, non-volatility and the list goes on. In this case, you cannot make any assumptions regarding the 'risk' of the financial asset relative to the market asset just from the sign of
.
Since beta is a measure of systematic risk, what would this mean by having a negative systematic risk?
Similarly, there is no universal agreement as to what kind of 'risk'
captures. Mathematically, it can be shown
fits into many models of 'risk' including VaR, CoVaR, MES, SRISK, DIP etc. Each of the theoretical papers that propose these models measure 'risk' from a different perspective. Empirically, there are evidence that
plays no role in diversification in emerging markets (there are a few seminal JF papers on this) and strangely enough,
sometimes has no explanatory power at all in FF3F or a momentum capturing model. So the result today is that
is just simply a mathematical measure of covariance; trying to generalize the meaning of
has been well documented in previous papers and there has not been any universal agreement (similar to all other areas of finance).
Furthermore, systematic risk itself is not well defined either. There are many models that attempt to capture "systematic risk" through the use of: Granger-causality tests, extreme value theory, logit regressions, GARCH time varying models, stochastic volatility processes. Some document the existence of systematic risk, some do not. Some define systematic risk in terms of skewness and standard deviation, some define it simply as a loss threshold. Thus, it is inappropriate to determine any relationships of
with systematic risk.