Using the data in the wikipedia link provided by @denesp we have the following picture, based on (2014 I think) Public Spending as percentage of GDP. The first column is GDP percentage bands, the second is the number of countries that fall in each band, the third column is the relative frequency, and the fourth the cumulative relative frequency. Total number of countries is $180$.
\begin{array}{| r | r | r | r|}
\hline
\text{%GDP range} & \text {count} & \text{rel. freq} & \text{Cum. rel. freq} \\
\hline
\hline
0-5 &0 & \text{0.0%} & \text{0.0%}
\\
\hline
5-10 & 0 & \text{0.0%}& \text{0.0%}
\\
\hline
10-15 & 0 & \text{0.0%} & \text{0.0%}\\
\hline
15-20 &18 & \text{10.0%} & \text{10.0%}\\
\hline
20-25 &26 &\text{14.4%} &\text{24.4%}
\\
\hline
25-30 &31 &\text{17.2%} &\text{41.6%}
\\
\hline
30-35 &26 &\text{14.4%} &\text{56.0%}
\\
\hline
35-40 &29 &\text{16.1%} &\text{72.1%}
\\
\hline
40-45 &17 &\text{9.4%} &\text{81.5%}
\\
\hline
45-50 &16 &\text{8.9%} &\text{90.4%}
\\
\hline
50-55 &9 &\text{5.0%} &\text{95.4%}
\\
\hline
55-60 &3 &\text{1.7%} &\text{97.1%}
\\
\hline
60-65 &1 &\text{0.6%} &\text{97.7%}
\\
\hline
65-70 &3 &\text{1.7%} &\text{99.4%}
\\
\hline
70-75 &0 &\text{0.0%} &\text{99.4%}
\\
\hline
75-80 &0 &\text{0.0%} &\text{99.4%}
\\
\hline
80-85 &0 &\text{0.0%} &\text{99.4%}
\\
\hline
85-90 &0 &\text{0.0%} &\text{99.4%}
\\
\hline
90-95 &1 &\text{0.6%} &\text{100.0%}
\\
\hline
95-100 &0 &\text{0.0%} &\text{100.0%}
\\
\hline
\end{array}
The data support a Gamma distribution with shape paramater $k=7.87$ and scale parameter $\theta = 4.29$, for which we have, ($X=$ Pubic spending as GDP %):

$$E(X) = 33.76 ,\;\; SD(X) = 12.3, median \approx 32.3$$
The empirical median is $32$ GDP percentage points.
Note that $90$% of the countries (163 entities in all) are below $50$% public spending as GDP percentage.