PostgreSQL - Optimizing query with multiple table selects

What would be the most optimized way to write the following multi select query based on a foreign key relationship?

SELECT
    a.id AS account,
    (SELECT COUNT(*) FROM b WHERE b.a_id = a.id) AS b_count, 
    (SELECT COUNT(*) FROM c WHERE c.a_id = a.id) AS c_count,
    (SELECT COUNT(*) FROM d WHERE d.a_id = a.id) AS d_count
FROM
    a

The query results should look like;

account| b_count | c_count | d_count  |
---------------------------------------
1      | 3325    | 5004    | 8002     |
---------------------------------------
2      | 52353   | 3009    | 1202     |

To be transparent the row volume for query will be roughly;

table a: 1,200 rows 
table b: 500,000 rows
table c: 600,000 rows
table d: 300,000 rows

Try this. I guess the performance will be better.
Perhaps the optimizer itself will make such a transformation of your request.
If you have index (a_id) for tables b,c,d - will be better.

select a.id
   ,coalesce(b_count,0)b_count
   ,coalesce(c_count,0)c_count
   ,coalesce(d_count,0)d_count
from a
left join( select a_id,count(*) b_count from b group by b.a_id)b on b.a_id=a.id
left join( select a_id,count(*) c_count from c group by c.a_id)c on c.a_id=a.id
left join( select a_id,count(*) d_count from d group by d.a_id)d on d.a_id=a.id

It will be interesting if you show a comparison of the performance of different queries. You have a large enough amount of data to test.