IN vs EXISTS

Queries with either IN or EXISTS is taking similar time. Why?
I have following 2 queries:declare
@ProductID as varchar(5) set
@ProductID = 'DS'IF @ProductID IN (select OrigProductCode from productsystemmap
where origProductCode
in (select distinct origproductcode from CTVMoneyMarketPortfolioProduct))
PRINT 'Y'ELSE
PRINT 'N'GOdeclare
@ProductID as varchar(5) set
@ProductID = 'DS'IF EXISTS (select OrigProductCode from productsystemmap MP WHERE OrigProductCode = @ProductID
AND EXISTS (select * from CTVMoneyMarketPortfolioProduct WHERE CTVPortfolioID IS NOT NULL AND origproductcode = MP.origproductcode))
PRINT 'Y'ELSE
PRINT 'N'
Where the table definition is as follows:GOCREATE
TABLE dbo.CTVMoneyMarketPortfolioProduct(CTVPortfolioID
int,OrigProductCode
char(8)
CONSTRAINT [PK__CTVPortfolio_Product] PRIMARY KEY CLUSTERED ([CTVPortfolioID]
ASC,[OrigProductCode]
ASC)
ON [PRIMARY]) ON [PRIMARY]
GOGOCREATE
TABLE [dbo].[ProductSystemMap]([OrigProductCode] [char]
(8) COLLATE Latin1_General_CI_AS NOT NULL,[ProductCode] [char]
(8) COLLATE Latin1_General_CI_AS NOT NULL,[OrigSystemID] [int]
NULL,
CONSTRAINT [PK_ProductSystemMap_1__16] PRIMARY KEY CLUSTERED ([OrigProductCode]
ASC)
ON [PRIMARY]) ON [PRIMARY]
GO
Both the tables do not have huge amount of data.
The execution plan is similar, with a Index Scan on CTVMoneyMarketPortfolioProduct. Is this because of composite primary key. How can i convert this index scan to index seek?

EXISTS will yield better performance than IN in case there are lots of matches for the subquery. So if the subquery reflects a one-to-many relationship, and there are multiple child records for most parent records, then EXISTS is probably faster. If you have only small amounts of data, you should not notice much of a difference anyway.
EXISTS will not perform worse than IN, so it is a safe bet.
... and to expand on that: NOT EXISTS will perform as poorly as NOT IN.

In addition to Adriaan's view, you can even compare the statistics of both queries by executing
SET STATISTICS IO ON
before your in or exist query on bigger table and check the scan count, logical reads etc.