Seaborn Catplot set values over the bars

It’s a rough solution but it does the trick.

We add the text to the axes object created by the plot.

The Y position is simple, as it corresponds exactly to the data value. We might just add 500 to each value so that the label sits nicely on top of the column.
The X position starts and is centered at 0 for the first group of columns (2013) and it spaces a unit. We have a buffer of 0.1 at each side and the columns are 5, hence each column is 0.16 wide.

g = sns.catplot(x='year', y='income', data=df, kind='bar', hue='geo_name', legend=True)
#flatax=g.axes.flatten()
#g.axes[0].text=('1')
g.fig.set_size_inches(15,8)
g.fig.subplots_adjust(top=0.81,right=0.86)
g.ax.text(-0.5,51000,'X=-0.5')
g.ax.text(-0.4,49000,'X=-0.4')
g.ax.text(0,49000,'X=0')
g.ax.text(0.5,51000,'X=0.5')
g.ax.text(0.4,49000,'X=0.4')
g.ax.text(0.6,47000,'X=0.6')

The text is, by default, aligned left (i.e. to the x value we set). Here is the documentation if you want to play with the text (change font, alignment, etc.)

We can then find the right placement for each label, knowing that the 3rd column of each group will always be centered on the unit (0,1,2,3,4).

g = sns.catplot(x='year', y='income', data=df, kind='bar', hue='geo_name', legend=True)
#flatax=g.axes.flatten()
#g.axes[0].text=('1')
g.fig.set_size_inches(15,8)
g.fig.subplots_adjust(top=0.81,right=0.86)
g.ax.text(-0.4,48411+500,'48,4K')
g.ax.text(-0.24,59055+500,'59,0K')
g.ax.text(-0.08,55003+500,'55,0K')
g.ax.text(0.08,37192+500,'37,2K')
g.ax.text(0.24,59055+500,'59,0K')

Of course, instead of manually labelling everything you should loop through the data and create the labels automatically

for i, yr in enumerate(df['year'].unique()):
    for j,gn in enumerate(df['geo_name'].unique()):

Now you can iterate through your x position using: i-0.4+(j*0.16) and at the same time you have the value for year and geo_name to retrieve the correct value of income.

We can use the Facet grid returned by sns.catplot() and select the axis. Use a for loop to position the Y-axis value in the format we need using ax.text()

g = sns.catplot(x='year', y='income', data=data, kind='bar', hue='geo_name', legend=True)
g.fig.set_size_inches(16,8)
g.fig.subplots_adjust(top=0.81,right=0.86) 

ax = g.facet_axis(0,0)
for p in ax.patches:
    ax.text(p.get_x() - 0.01, 
            p.get_height() * 1.02, 
           '{0:.1f}K'.format(p.get_height()/1000),   #Used to format it K representation
            color='black', 
            rotation='horizontal', 
            size='large')